Abstract
We have developed a comprehensive computer system to assist farmers who practice traditional farming methods and have limited access to agricultural experts for addressing crop diseases. Our system utilizes artificial intelligence (AI) to identify and provide remedies for vegetable diseases. To ensure ease of use, we have created a mobile application that offers a user-friendly interface, allowing farmers to inquire about vegetable diseases and receive suitable solutions in their local language. The developed system can be utilized by any farmer with a basic understanding of a smartphone. Specifically, we have designed an AI-enabled mobile application for identifying and suggesting remedies for vegetable diseases, focusing on tomato diseases to benefit the local farming community in Nepal. Our system employs state-of-the-art object detection methodology, namely You Only Look Once (YOLO), to detect tomato diseases. The detected information is then relayed to the mobile application, which provides remedy suggestions guided by domain experts. In order to train our system effectively, we curated a dataset consisting of ten classes of tomato diseases. We utilized various data augmentation methods to address overfitting and trained a YOLOv5 object detector. The proposed method achieved a mean average precision of 0.76 and offers an efficient mobile interface for interacting with the AI system. While our system is currently in the development phase, we are actively working towards enhancing its robustness and real-time usability by accumulating more training samples.
Robust and Efficient Interference Neural Networks for Defending Against Adversarial Attacks in ImageNet
Authors: Yunuo Xiong, Shujuan Liu, Hongwei Xiong
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Abstract
The existence of adversarial images has seriously affected the task of image recognition and practical application of deep learning, it is also a key scientific problem that deep learning urgently needs to solve. By far the most effective approach is to train the neural network with a large number of adversarial examples. However, this adversarial training method requires a huge amount of computing resources when applied to ImageNet, and has not yet achieved satisfactory results for high-intensity adversarial attacks. In this paper, we construct an interference neural network by applying additional background images and corresponding labels, and use pre-trained ResNet-152 to efficiently complete the training. Compared with the state-of-the-art results under the PGD attack, it has a better defense effect with much smaller computing resources. This work provides new ideas for academic research and practical applications of effective defense against adversarial attacks.
Reducing the False Positive Rate Using Bayesian Inference in Autonomous Driving Perception
Authors: Johann J. S. Bastos, Bruno L. S. da Silva, Tiago Zanotelli, Cristiano Premebida, Gledson Melotti
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
Object recognition is a crucial step in perception systems for autonomous and intelligent vehicles, as evidenced by the numerous research works in the topic. In this paper, object recognition is explored by using multisensory and multimodality approaches, with the intention of reducing the false positive rate (FPR). The reduction of the FPR becomes increasingly important in perception systems since the misclassification of an object can potentially cause accidents. In particular, this work presents a strategy through Bayesian inference to reduce the FPR considering the likelihood function as a cumulative distribution function from Gaussian kernel density estimations, and the prior probabilities as cumulative functions of normalized histograms. The validation of the proposed methodology is performed on the KITTI dataset using deep networks (DenseNet, NasNet, and EfficientNet), and recent 3D point cloud networks (PointNet, and PintNet++), by considering three object-categories (cars, cyclists, pedestrians) and the RGB and LiDAR sensor modalities.
DynamicBEV: Leveraging Dynamic Queries and Temporal Context for 3D Object Detection
Authors: Jiawei Yao, Yingxin Lai
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
3D object detection is crucial for applications like autonomous driving and robotics. While query-based 3D object detection for BEV (Bird's Eye View) images has seen significant advancements, most existing methods follows the paradigm of static query. Such paradigm is incapable of adapting to complex spatial-temporal relationships in the scene. To solve this problem, we introduce a new paradigm in DynamicBEV, a novel approach that employs dynamic queries for BEV-based 3D object detection. In contrast to static queries, the proposed dynamic queries exploit K-means clustering and Top-K Attention in a creative way to aggregate information more effectively from both local and distant feature, which enable DynamicBEV to adapt iteratively to complex scenes. To further boost efficiency, DynamicBEV incorporates a Lightweight Temporal Fusion Module (LTFM), designed for efficient temporal context integration with a significant computation reduction. Additionally, a custom-designed Diversity Loss ensures a balanced feature representation across scenarios. Extensive experiments on the nuScenes dataset validate the effectiveness of DynamicBEV, establishing a new state-of-the-art and heralding a paradigm-level breakthrough in query-based BEV object detection.
LCOT: Linear circular optimal transport
Authors: Rocio Diaz Martin, Ivan Medri, Yikun Bai, Xinran Liu, Kangbai Yan, Gustavo K. Rohde, Soheil Kolouri
Abstract
The optimal transport problem for measures supported on non-Euclidean spaces has recently gained ample interest in diverse applications involving representation learning. In this paper, we focus on circular probability measures, i.e., probability measures supported on the unit circle, and introduce a new computationally efficient metric for these measures, denoted as Linear Circular Optimal Transport (LCOT). The proposed metric comes with an explicit linear embedding that allows one to apply Machine Learning (ML) algorithms to the embedded measures and seamlessly modify the underlying metric for the ML algorithm to LCOT. We show that the proposed metric is rooted in the Circular Optimal Transport (COT) and can be considered the linearization of the COT metric with respect to a fixed reference measure. We provide a theoretical analysis of the proposed metric and derive the computational complexities for pairwise comparison of circular probability measures. Lastly, through a set of numerical experiments, we demonstrate the benefits of LCOT in learning representations of circular measures.
Abstract
As the ubiquity and complexity of system-on-chip (SoC) designs increase across electronic devices, the task of incorporating security into an SoC design flow poses significant challenges. Existing security solutions are inadequate to provide effective verification of modern SoC designs due to their limitations in scalability, comprehensiveness, and adaptability. On the other hand, Large Language Models (LLMs) are celebrated for their remarkable success in natural language understanding, advanced reasoning, and program synthesis tasks. Recognizing an opportunity, our research delves into leveraging the emergent capabilities of Generative Pre-trained Transformers (GPTs) to address the existing gaps in SoC security, aiming for a more efficient, scalable, and adaptable methodology. By integrating LLMs into the SoC security verification paradigm, we open a new frontier of possibilities and challenges to ensure the security of increasingly complex SoCs. This paper offers an in-depth analysis of existing works, showcases practical case studies, demonstrates comprehensive experiments, and provides useful promoting guidelines. We also present the achievements, prospects, and challenges of employing LLM in different SoC security verification tasks.
Towards Agility: A Momentum Aware Trajectory Optimisation Framework using Full-Centroidal Dynamics & Implicit Inverse Kinematics
Authors: Aristotelis Papatheodorou, Wolfgang Merkt, Alexander L. Mitchell, Ioannis Havoutis
Abstract
Online planning and execution of acrobatic maneuvers pose significant challenges in legged locomotion. Their underlying combinatorial nature, along with the current hardware's limitations constitute the main obstacles in unlocking the true potential of legged-robots. This letter tries to expose the intricacies of these optimal control problems in a tangible way, directly applicable to the creation of more efficient online trajectory optimisation frameworks. By analysing the fundamental principles that shape the behaviour of the system, the dynamics themselves can be exploited to surpass its hardware limitations. More specifically, a trajectory optimisation formulation is proposed that exploits the system's high-order nonlinearities, such as the nonholonomy of the angular momentum, and phase-space symmetries in order to produce feasible high-acceleration maneuvers. By leveraging the full-centroidal dynamics of the quadruped ANYmal C and directly optimising its footholds and contact forces, the framework is capable of producing efficient motion plans with low computational overhead. The feasibility of the produced trajectories is ensured by taking into account the configuration-dependent inertial properties of the robot during the planning process, while its robustness is increased by supplying the full analytic derivatives & hessians to the solver. Finally, a significant portion of the discussion is centred around the deployment of the proposed framework on the ANYmal C platform, while its true capabilities are demonstrated through real-world experiments, with the successful execution of high-acceleration motion scenarios like the squat-jump.
Numerical Analysis of time-dependent Hamilton-Jacobi Equations on Networks
Authors: Elisabetta Carlini, Antonio Siconolfi
Subjects: Numerical Analysis (math.NA); Analysis of PDEs (math.AP)
Abstract
A new algorithm for time dependent Hamilton Jacobi equations on networks, based on semi Lagrangian scheme, is proposed. It is based on the definition of viscosity solution for this kind of problems recently given in. A thorough convergence analysis, not requiring weak semilimits, is provided. In particular, the check of the supersolution property at the vertices is performed through a dynamical technique which seems new. The scheme is efficient, explicit, allows long time steps, and is suitable to be implemented in a parallel algorithm. We present some numerical tests, showing the advantage in terms of computational cost over the one proposed in [7]
QR-Tag: Angular Measurement and Tracking with a QR-Design Marker
Authors: Simeng Qiu, Hadi Amata, Wolfgang Heidrich
Subjects: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
Abstract
Directional information measurement has many applications in domains such as robotics, virtual and augmented reality, and industrial computer vision. Conventional methods either require pre-calibration or necessitate controlled environments. The state-of-the-art MoireTag approach exploits the Moire effect and QR-design to continuously track the angular shift precisely. However, it is still not a fully QR code design. To overcome the above challenges, we propose a novel snapshot method for discrete angular measurement and tracking with scannable QR-design patterns that are generated by binary structures printed on both sides of a glass plate. The QR codes, resulting from the parallax effect due to the geometry alignment between two layers, can be readily measured as angular information using a phone camera. The simulation results show that the proposed non-contact object tracking framework is computationally efficient with high accuracy.
When is Agnostic Reinforcement Learning Statistically Tractable?
Abstract
We study the problem of agnostic PAC reinforcement learning (RL): given a policy class $\Pi$, how many rounds of interaction with an unknown MDP (with a potentially large state and action space) are required to learn an $\epsilon$-suboptimal policy with respect to $\Pi$? Towards that end, we introduce a new complexity measure, called the \emph{spanning capacity}, that depends solely on the set $\Pi$ and is independent of the MDP dynamics. With a generative model, we show that for any policy class $\Pi$, bounded spanning capacity characterizes PAC learnability. However, for online RL, the situation is more subtle. We show there exists a policy class $\Pi$ with a bounded spanning capacity that requires a superpolynomial number of samples to learn. This reveals a surprising separation for agnostic learnability between generative access and online access models (as well as between deterministic/stochastic MDPs under online access). On the positive side, we identify an additional \emph{sunflower} structure, which in conjunction with bounded spanning capacity enables statistically efficient online RL via a new algorithm called POPLER, which takes inspiration from classical importance sampling methods as well as techniques for reachable-state identification and policy evaluation in reward-free exploration.
On Time Domain Conformer Models for Monaural Speech Separation in Noisy Reverberant Acoustic Environments
Authors: William Ravenscroft, Stefan Goetze, Thomas Hain
Abstract
Speech separation remains an important topic for multi-speaker technology researchers. Convolution augmented transformers (conformers) have performed well for many speech processing tasks but have been under-researched for speech separation. Most recent state-of-the-art (SOTA) separation models have been time-domain audio separation networks (TasNets). A number of successful models have made use of dual-path (DP) networks which sequentially process local and global information. Time domain conformers (TD-Conformers) are an analogue of the DP approach in that they also process local and global context sequentially but have a different time complexity function. It is shown that for realistic shorter signal lengths, conformers are more efficient when controlling for feature dimension. Subsampling layers are proposed to further improve computational efficiency. The best TD-Conformer achieves 14.6 dB and 21.2 dB SISDR improvement on the WHAMR and WSJ0-2Mix benchmarks, respectively.
Learning Layer-wise Equivariances Automatically using Gradients
Authors: Tycho F.A. van der Ouderaa, Alexander Immer, Mark van der Wilk
Abstract
Convolutions encode equivariance symmetries into neural networks leading to better generalisation performance. However, symmetries provide fixed hard constraints on the functions a network can represent, need to be specified in advance, and can not be adapted. Our goal is to allow flexible symmetry constraints that can automatically be learned from data using gradients. Learning symmetry and associated weight connectivity structures from scratch is difficult for two reasons. First, it requires efficient and flexible parameterisations of layer-wise equivariances. Secondly, symmetries act as constraints and are therefore not encouraged by training losses measuring data fit. To overcome these challenges, we improve parameterisations of soft equivariance and learn the amount of equivariance in layers by optimising the marginal likelihood, estimated using differentiable Laplace approximations. The objective balances data fit and model complexity enabling layer-wise symmetry discovery in deep networks. We demonstrate the ability to automatically learn layer-wise equivariances on image classification tasks, achieving equivalent or improved performance over baselines with hard-coded symmetry.
Predicting Player Engagement in Tom Clancy's The Division 2: A Multimodal Approach via Pixels and Gamepad Actions
Authors: Kosmas Pinitas, David Renaudie, Mike Thomsen, Matthew Barthet, Konstantinos Makantasis, Antonios Liapis, Georgios N. Yannakakis
Abstract
This paper introduces a large scale multimodal corpus collected for the purpose of analysing and predicting player engagement in commercial-standard games. The corpus is solicited from 25 players of the action role-playing game Tom Clancy's The Division 2, who annotated their level of engagement using a time-continuous annotation tool. The cleaned and processed corpus presented in this paper consists of nearly 20 hours of annotated gameplay videos accompanied by logged gamepad actions. We report preliminary results on predicting long-term player engagement based on in-game footage and game controller actions using Convolutional Neural Network architectures. Results obtained suggest we can predict the player engagement with up to 72% accuracy on average (88% at best) when we fuse information from the game footage and the player's controller input. Our findings validate the hypothesis that long-term (i.e. 1 hour of play) engagement can be predicted efficiently solely from pixels and gamepad actions.
On the Correlation between Random Variables and their Principal Components
Authors: Zenon Gniazdowski
Subjects: Machine Learning (cs.LG); Statistics Theory (math.ST)
Abstract
The article attempts to find an algebraic formula describing the correlation coefficients between random variables and the principal components representing them. As a result of the analysis, starting from selected statistics relating to individual random variables, the equivalents of these statistics relating to a set of random variables were presented in the language of linear algebra, using the concepts of vector and matrix. This made it possible, in subsequent steps, to derive the expected formula. The formula found is identical to the formula used in Factor Analysis to calculate factor loadings. The discussion showed that it is possible to apply this formula to optimize the number of principal components in Principal Component Analysis, as well as to optimize the number of factors in Factor Analysis.
Entropy Based Multi-robot Active SLAM
Authors: Muhammad Farhan Ahmed, Matteo Maragliano, Vincent Frémont, Carmine Tommaso Recchiuto
Abstract
In this article, we present an efficient multi-robot active SLAM framework that involves a frontier-sharing method for maximum exploration of an unknown environment. It encourages the robots to spread into the environment while weighting the goal frontiers with the pose graph SLAM uncertainly and path entropy. Our approach works on a limited number of frontier points and weights the goal frontiers with a utility function that encapsulates both the SLAM and map uncertainties, thus providing an efficient and not computationally expensive solution. Our approach has been tested on publicly available simulation environments and on real robots. An accumulative 31% more coverage than similar state-of-the-art approaches has been obtained, proving the capability of our approach for efficient environment exploration.
Abstract
State-of-the-art coreference resolutions systems depend on multiple LLM calls per document and are thus prohibitively expensive for many use cases (e.g., information extraction with large corpora). The leading word-level coreference system (WL-coref) attains 96.6% of these SOTA systems' performance while being much more efficient. In this work, we identify a routine yet important failure case of WL-coref: dealing with conjoined mentions such as 'Tom and Mary'. We offer a simple yet effective solution that improves the performance on the OntoNotes test set by 0.9% F1, shrinking the gap between efficient word-level coreference resolution and expensive SOTA approaches by 34.6%. Our Conjunction-Aware Word-level coreference model (CAW-coref) and code is available at https://github.com/KarelDO/wl-coref.
Look-Up mAI GeMM: Increasing AI GeMMs Performance by Nearly 2.5x via msGeMM
Abstract
AI models are increasing in size and recent advancement in the community has shown that unlike HPC applications where double precision datatype are required, lower-precision datatypes such as fp8 or int4 are sufficient to bring the same model quality both for training and inference. Following these trends, GPU vendors such as NVIDIA and AMD have added hardware support for fp16, fp8 and int8 GeMM operations with an exceptional performance via Tensor Cores. However, this paper proposes a new algorithm called msGeMM which shows that AI models with low-precision datatypes can run with ~2.5x fewer multiplication and add instructions. Efficient implementation of this algorithm requires special CUDA cores with the ability to add elements from a small look-up table at the rate of Tensor Cores.
Automatic Integration for Spatiotemporal Neural Point Processes
Abstract
Learning continuous-time point processes is essential to many discrete event forecasting tasks. However, integration poses a major challenge, particularly for spatiotemporal point processes (STPPs), as it involves calculating the likelihood through triple integrals over space and time. Existing methods for integrating STPP either assume a parametric form of the intensity function, which lacks flexibility; or approximating the intensity with Monte Carlo sampling, which introduces numerical errors. Recent work by Omi et al. [2019] proposes a dual network or AutoInt approach for efficient integration of flexible intensity function. However, the method only focuses on the 1D temporal point process. In this paper, we introduce a novel paradigm: AutoSTPP (Automatic Integration for Spatiotemporal Neural Point Processes) that extends the AutoInt approach to 3D STPP. We show that direct extension of the previous work overly constrains the intensity function, leading to poor performance. We prove consistency of AutoSTPP and validate it on synthetic data and benchmark real world datasets, showcasing its significant advantage in recovering complex intensity functions from irregular spatiotemporal events, particularly when the intensity is sharply localized.
Quasi-Monte Carlo sparse grid Galerkin finite element methods for linear elasticity equations with uncertainties
Authors: J. Dick, Q. T. Le Gia, K. Mustapha, T. Tran
Abstract
We explore a linear inhomogeneous elasticity equation with random Lam\'e parameters. The latter are parameterized by a countably infinite number of terms in separated expansions. The main aim of this work is to estimate expected values (considered as an infinite dimensional integral on the parametric space corresponding to the random coefficients) of linear functionals acting on the solution of the elasticity equation. To achieve this, the expansions of the random parameters are truncated, a high-order quasi-Monte Carlo (QMC) is combined with a sparse grid approach to approximate the high dimensional integral, and a Galerkin finite element method (FEM) is introduced to approximate the solution of the elasticity equation over the physical domain. The error estimates from (1) truncating the infinite expansion, (2) the Galerkin FEM, and (3) the QMC sparse grid quadrature rule are all studied. For this purpose, we show certain required regularity properties of the continuous solution with respect to both the parametric and physical variables. To achieve our theoretical regularity and convergence results, some reasonable assumptions on the expansions of the random coefficients are imposed. Finally, some numerical results are delivered.
Motion Memory: Leveraging Past Experiences to Accelerate Future Motion Planning
Abstract
When facing a new motion-planning problem, most motion planners solve it from scratch, e.g., via sampling and exploration or starting optimization from a straight-line path. However, most motion planners have to experience a variety of planning problems throughout their lifetimes, which are yet to be leveraged for future planning. In this paper, we present a simple but efficient method called Motion Memory, which allows different motion planners to accelerate future planning using past experiences. Treating existing motion planners as either a closed or open box, we present a variety of ways that Motion Memory can contribute to reduce the planning time when facing a new planning problem. We provide extensive experiment results with three different motion planners on three classes of planning problems with over 30,000 problem instances and show that planning speed can be significantly reduced by up to 89% with the proposed Motion Memory technique and with increasing past planning experiences.
CAT-RRT: Motion Planning that Admits Contact One Link at a Time
Abstract
Current motion planning approaches rely on binary collision checking to evaluate the validity of a state and thereby dictate where the robot is allowed to move. This approach leaves little room for robots to engage in contact with an object, as is often necessary when operating in densely cluttered spaces. In this work, we propose an alternative method that considers contact states as high-cost states that the robot should avoid but can traverse if necessary to complete a task. More specifically, we introduce Contact Admissible Transition-based Rapidly exploring Random Trees (CAT-RRT), a planner that uses a novel per-link cost heuristic to find a path by traversing high-cost obstacle regions. Through extensive testing, we find that state-of-the-art optimization planners tend to over-explore low-cost states, which leads to slow and inefficient convergence to contact regions. Conversely, CAT-RRT searches both low and high-cost regions simultaneously with an adaptive thresholding mechanism carried out at each robot link. This leads to paths with a balance between efficiency, path length, and contact cost.
GeoLLM: Extracting Geospatial Knowledge from Large Language Models
Authors: Rohin Manvi, Samar Khanna, Gengchen Mai, Marshall Burke, David Lobell, Stefano Ermon
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Abstract
The application of machine learning (ML) in a range of geospatial tasks is increasingly common but often relies on globally available covariates such as satellite imagery that can either be expensive or lack predictive power. Here we explore the question of whether the vast amounts of knowledge found in Internet language corpora, now compressed within large language models (LLMs), can be leveraged for geospatial prediction tasks. We first demonstrate that LLMs embed remarkable spatial information about locations, but naively querying LLMs using geographic coordinates alone is ineffective in predicting key indicators like population density. We then present GeoLLM, a novel method that can effectively extract geospatial knowledge from LLMs with auxiliary map data from OpenStreetMap. We demonstrate the utility of our approach across multiple tasks of central interest to the international community, including the measurement of population density and economic livelihoods. Across these tasks, our method demonstrates a 70% improvement in performance (measured using Pearson's $r^2$) relative to baselines that use nearest neighbors or use information directly from the prompt, and performance equal to or exceeding satellite-based benchmarks in the literature. With GeoLLM, we observe that GPT-3.5 outperforms Llama 2 and RoBERTa by 19% and 51% respectively, suggesting that the performance of our method scales well with the size of the model and its pretraining dataset. Our experiments reveal that LLMs are remarkably sample-efficient, rich in geospatial information, and robust across the globe. Crucially, GeoLLM shows promise in mitigating the limitations of existing geospatial covariates and complementing them well.
CoT3DRef: Chain-of-Thoughts Data-Efficient 3D Visual Grounding
Abstract
3D visual grounding is the ability to localize objects in 3D scenes conditioned by utterances. Most existing methods devote the referring head to localize the referred object directly, causing failure in complex scenarios. In addition, it does not illustrate how and why the network reaches the final decision. In this paper, we address this question Can we design an interpretable 3D visual grounding framework that has the potential to mimic the human perception system?. To this end, we formulate the 3D visual grounding problem as a sequence-to-sequence task by first predicting a chain of anchors and then the final target. Interpretability not only improves the overall performance but also helps us identify failure cases. Following the chain of thoughts approach enables us to decompose the referring task into interpretable intermediate steps, boosting the performance and making our framework extremely data-efficient. Moreover, our proposed framework can be easily integrated into any existing architecture. We validate our approach through comprehensive experiments on the Nr3D, Sr3D, and Scanrefer benchmarks and show consistent performance gains compared to existing methods without requiring manually annotated data. Furthermore, our proposed framework, dubbed CoT3DRef, is significantly data-efficient, whereas on the Sr3D dataset, when trained only on 10% of the data, we match the SOTA performance that trained on the entire data.
Spiking PointNet: Spiking Neural Networks for Point Clouds
Abstract
Recently, Spiking Neural Networks (SNNs), enjoying extreme energy efficiency, have drawn much research attention on 2D visual recognition and shown gradually increasing application potential. However, it still remains underexplored whether SNNs can be generalized to 3D recognition. To this end, we present Spiking PointNet in the paper, the first spiking neural model for efficient deep learning on point clouds. We discover that the two huge obstacles limiting the application of SNNs in point clouds are: the intrinsic optimization obstacle of SNNs that impedes the training of a big spiking model with large time steps, and the expensive memory and computation cost of PointNet that makes training a big spiking point model unrealistic. To solve the problems simultaneously, we present a trained-less but learning-more paradigm for Spiking PointNet with theoretical justifications and in-depth experimental analysis. In specific, our Spiking PointNet is trained with only a single time step but can obtain better performance with multiple time steps inference, compared to the one trained directly with multiple time steps. We conduct various experiments on ModelNet10, ModelNet40 to demonstrate the effectiveness of Spiking PointNet. Notably, our Spiking PointNet even can outperform its ANN counterpart, which is rare in the SNN field thus providing a potential research direction for the following work. Moreover, Spiking PointNet shows impressive speedup and storage saving in the training phase.
Low-Rank Tensor Completion via Novel Sparsity-Inducing Regularizers
Authors: Zhi-Yong Wang, Hing Cheung So, Abdelhak M. Zoubir
Subjects: Machine Learning (cs.LG); Image and Video Processing (eess.IV); Signal Processing (eess.SP)
Abstract
To alleviate the bias generated by the l1-norm in the low-rank tensor completion problem, nonconvex surrogates/regularizers have been suggested to replace the tensor nuclear norm, although both can achieve sparsity. However, the thresholding functions of these nonconvex regularizers may not have closed-form expressions and thus iterations are needed, which increases the computational loads. To solve this issue, we devise a framework to generate sparsity-inducing regularizers with closed-form thresholding functions. These regularizers are applied to low-tubal-rank tensor completion, and efficient algorithms based on the alternating direction method of multipliers are developed. Furthermore, convergence of our methods is analyzed and it is proved that the generated sequences are bounded and any limit point is a stationary point. Experimental results using synthetic and real-world datasets show that the proposed algorithms outperform the state-of-the-art methods in terms of restoration performance.
Efficient Adaptation of Large Vision Transformer via Adapter Re-Composing
Authors: Wei Dong, Dawei Yan, Zhijun Lin, Peng Wang
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Abstract
The advent of high-capacity pre-trained models has revolutionized problem-solving in computer vision, shifting the focus from training task-specific models to adapting pre-trained models. Consequently, effectively adapting large pre-trained models to downstream tasks in an efficient manner has become a prominent research area. Existing solutions primarily concentrate on designing lightweight adapters and their interaction with pre-trained models, with the goal of minimizing the number of parameters requiring updates. In this study, we propose a novel Adapter Re-Composing (ARC) strategy that addresses efficient pre-trained model adaptation from a fresh perspective. Our approach considers the reusability of adaptation parameters and introduces a parameter-sharing scheme. Specifically, we leverage symmetric down-/up-projections to construct bottleneck operations, which are shared across layers. By learning low-dimensional re-scaling coefficients, we can effectively re-compose layer-adaptive adapters. This parameter-sharing strategy in adapter design allows us to significantly reduce the number of new parameters while maintaining satisfactory performance, thereby offering a promising approach to compress the adaptation cost. We conduct experiments on 24 downstream image classification tasks using various Vision Transformer variants to evaluate our method. The results demonstrate that our approach achieves compelling transfer learning performance with a reduced parameter count. Our code is available at \href{https://github.com/DavidYanAnDe/ARC}{https://github.com/DavidYanAnDe/ARC}.
Sample-Efficient Multi-Agent RL: An Optimization Perspective
Authors: Nuoya Xiong, Zhihan Liu, Zhaoran Wang, Zhuoran Yang
Subjects: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT)
Abstract
We study multi-agent reinforcement learning (MARL) for the general-sum Markov Games (MGs) under the general function approximation. In order to find the minimum assumption for sample-efficient learning, we introduce a novel complexity measure called the Multi-Agent Decoupling Coefficient (MADC) for general-sum MGs. Using this measure, we propose the first unified algorithmic framework that ensures sample efficiency in learning Nash Equilibrium, Coarse Correlated Equilibrium, and Correlated Equilibrium for both model-based and model-free MARL problems with low MADC. We also show that our algorithm provides comparable sublinear regret to the existing works. Moreover, our algorithm combines an equilibrium-solving oracle with a single objective optimization subprocedure that solves for the regularized payoff of each deterministic joint policy, which avoids solving constrained optimization problems within data-dependent constraints (Jin et al. 2020; Wang et al. 2023) or executing sampling procedures with complex multi-objective optimization problems (Foster et al. 2023), thus being more amenable to empirical implementation.
A Unified View on Solving Objective Mismatch in Model-Based Reinforcement Learning
Authors: Ran Wei, Nathan Lambert, Anthony McDonald, Alfredo Garcia, Roberto Calandra
Abstract
Model-based Reinforcement Learning (MBRL) aims to make agents more sample-efficient, adaptive, and explainable by learning an explicit model of the environment. While the capabilities of MBRL agents have significantly improved in recent years, how to best learn the model is still an unresolved question. The majority of MBRL algorithms aim at training the model to make accurate predictions about the environment and subsequently using the model to determine the most rewarding actions. However, recent research has shown that model predictive accuracy is often not correlated with action quality, tracing the root cause to the \emph{objective mismatch} between accurate dynamics model learning and policy optimization of rewards. A number of interrelated solution categories to the objective mismatch problem have emerged as MBRL continues to mature as a research area. In this work, we provide an in-depth survey of these solution categories and propose a taxonomy to foster future research.
Bi-Level Offline Policy Optimization with Limited Exploration
Authors: Wenzhuo Zhou
Subjects: Machine Learning (cs.LG); Statistics Theory (math.ST)
Abstract
We study offline reinforcement learning (RL) which seeks to learn a good policy based on a fixed, pre-collected dataset. A fundamental challenge behind this task is the distributional shift due to the dataset lacking sufficient exploration, especially under function approximation. To tackle this issue, we propose a bi-level structured policy optimization algorithm that models a hierarchical interaction between the policy (upper-level) and the value function (lower-level). The lower level focuses on constructing a confidence set of value estimates that maintain sufficiently small weighted average Bellman errors, while controlling uncertainty arising from distribution mismatch. Subsequently, at the upper level, the policy aims to maximize a conservative value estimate from the confidence set formed at the lower level. This novel formulation preserves the maximum flexibility of the implicitly induced exploratory data distribution, enabling the power of model extrapolation. In practice, it can be solved through a computationally efficient, penalized adversarial estimation procedure. Our theoretical regret guarantees do not rely on any data-coverage and completeness-type assumptions, only requiring realizability. These guarantees also demonstrate that the learned policy represents the "best effort" among all policies, as no other policies can outperform it. We evaluate our model using a blend of synthetic, benchmark, and real-world datasets for offline RL, showing that it performs competitively with state-of-the-art methods.
Towards More Efficient Depression Risk Recognition via Gait
Abstract
Depression, a highly prevalent mental illness, affects over 280 million individuals worldwide. Early detection and timely intervention are crucial for promoting remission, preventing relapse, and alleviating the emotional and financial burdens associated with depression. However, patients with depression often go undiagnosed in the primary care setting. Unlike many physiological illnesses, depression lacks objective indicators for recognizing depression risk, and existing methods for depression risk recognition are time-consuming and often encounter a shortage of trained medical professionals. The correlation between gait and depression risk has been empirically established. Gait can serve as a promising objective biomarker, offering the advantage of efficient and convenient data collection. However, current methods for recognizing depression risk based on gait have only been validated on small, private datasets, lacking large-scale publicly available datasets for research purposes. Additionally, these methods are primarily limited to hand-crafted approaches. Gait is a complex form of motion, and hand-crafted gait features often only capture a fraction of the intricate associations between gait and depression risk. Therefore, this study first constructs a large-scale gait database, encompassing over 1,200 individuals, 40,000 gait sequences, and covering six perspectives and three types of attire. Two commonly used psychological scales are provided as depression risk annotations. Subsequently, a deep learning-based depression risk recognition model is proposed, overcoming the limitations of hand-crafted approaches. Through experiments conducted on the constructed large-scale database, the effectiveness of the proposed method is validated, and numerous instructive insights are presented in the paper, highlighting the significant potential of gait-based depression risk recognition.
Dynamical versus Bayesian Phase Transitions in a Toy Model of Superposition
Authors: Zhongtian Chen, Edmund Lau, Jake Mendel, Susan Wei, Daniel Murfet
Abstract
We investigate phase transitions in a Toy Model of Superposition (TMS) using Singular Learning Theory (SLT). We derive a closed formula for the theoretical loss and, in the case of two hidden dimensions, discover that regular $k$-gons are critical points. We present supporting theory indicating that the local learning coefficient (a geometric invariant) of these $k$-gons determines phase transitions in the Bayesian posterior as a function of training sample size. We then show empirically that the same $k$-gon critical points also determine the behavior of SGD training. The picture that emerges adds evidence to the conjecture that the SGD learning trajectory is subject to a sequential learning mechanism. Specifically, we find that the learning process in TMS, be it through SGD or Bayesian learning, can be characterized by a journey through parameter space from regions of high loss and low complexity to regions of low loss and high complexity.
Learning bounded-degree polytrees with known skeleton
Authors: Davin Choo, Joy Qiping Yang, Arnab Bhattacharyya, Clément L. Canonne
Subjects: Machine Learning (cs.LG); Data Structures and Algorithms (cs.DS); Probability (math.PR); Statistics Theory (math.ST); Machine Learning (stat.ML)
Abstract
We establish finite-sample guarantees for efficient proper learning of bounded-degree polytrees, a rich class of high-dimensional probability distributions and a subclass of Bayesian networks, a widely-studied type of graphical model. Recently, Bhattacharyya et al. (2021) obtained finite-sample guarantees for recovering tree-structured Bayesian networks, i.e., 1-polytrees. We extend their results by providing an efficient algorithm which learns $d$-polytrees in polynomial time and sample complexity for any bounded $d$ when the underlying undirected graph (skeleton) is known. We complement our algorithm with an information-theoretic sample complexity lower bound, showing that the dependence on the dimension and target accuracy parameters are nearly tight.
Boosting Continuous Control with Consistency Policy
Abstract
Due to its training stability and strong expression, the diffusion model has attracted considerable attention in offline reinforcement learning. However, several challenges have also come with it: 1) The demand for a large number of diffusion steps makes the diffusion-model-based methods time inefficient and limits their applications in real-time control; 2) How to achieve policy improvement with accurate guidance for diffusion model-based policy is still an open problem. Inspired by the consistency model, we propose a novel time-efficiency method named Consistency Policy with Q-Learning (CPQL), which derives action from noise by a single step. By establishing a mapping from the reverse diffusion trajectories to the desired policy, we simultaneously address the issues of time efficiency and inaccurate guidance when updating diffusion model-based policy with the learned Q-function. We demonstrate that CPQL can achieve policy improvement with accurate guidance for offline reinforcement learning, and can be seamlessly extended for online RL tasks. Experimental results indicate that CPQL achieves new state-of-the-art performance on 11 offline and 21 online tasks, significantly improving inference speed by nearly 45 times compared to Diffusion-QL. We will release our code later.
Filter Pruning For CNN With Enhanced Linear Representation Redundancy
Abstract
Structured network pruning excels non-structured methods because they can take advantage of the thriving developed parallel computing techniques. In this paper, we propose a new structured pruning method. Firstly, to create more structured redundancy, we present a data-driven loss function term calculated from the correlation coefficient matrix of different feature maps in the same layer, named CCM-loss. This loss term can encourage the neural network to learn stronger linear representation relations between feature maps during the training from the scratch so that more homogenous parts can be removed later in pruning. CCM-loss provides us with another universal transcendental mathematical tool besides L*-norm regularization, which concentrates on generating zeros, to generate more redundancy but for the different genres. Furthermore, we design a matching channel selection strategy based on principal components analysis to exploit the maximum potential ability of CCM-loss. In our new strategy, we mainly focus on the consistency and integrality of the information flow in the network. Instead of empirically hard-code the retain ratio for each layer, our channel selection strategy can dynamically adjust each layer's retain ratio according to the specific circumstance of a per-trained model to push the prune ratio to the limit. Notably, on the Cifar-10 dataset, our method brings 93.64% accuracy for pruned VGG-16 with only 1.40M parameters and 49.60M FLOPs, the pruned ratios for parameters and FLOPs are 90.6% and 84.2%, respectively. For ResNet-50 trained on the ImageNet dataset, our approach achieves 42.8% and 47.3% storage and computation reductions, respectively, with an accuracy of 76.23%. Our code is available at https://github.com/Bojue-Wang/CCM-LRR.
JointNet: Extending Text-to-Image Diffusion for Dense Distribution Modeling
Authors: Jingyang Zhang, Shiwei Li, Yuanxun Lu, Tian Fang, David McKinnon, Yanghai Tsin, Long Quan, Yao Yao
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
We introduce JointNet, a novel neural network architecture for modeling the joint distribution of images and an additional dense modality (e.g., depth maps). JointNet is extended from a pre-trained text-to-image diffusion model, where a copy of the original network is created for the new dense modality branch and is densely connected with the RGB branch. The RGB branch is locked during network fine-tuning, which enables efficient learning of the new modality distribution while maintaining the strong generalization ability of the large-scale pre-trained diffusion model. We demonstrate the effectiveness of JointNet by using RGBD diffusion as an example and through extensive experiments, showcasing its applicability in a variety of applications, including joint RGBD generation, dense depth prediction, depth-conditioned image generation, and coherent tile-based 3D panorama generation.
Advanced Efficient Strategy for Detection of Dark Objects Based on Spiking Network with Multi-Box Detection
Abstract
Several deep learning algorithms have shown amazing performance for existing object detection tasks, but recognizing darker objects is the largest challenge. Moreover, those techniques struggled to detect or had a slow recognition rate, resulting in significant performance losses. As a result, an improved and accurate detection approach is required to address the above difficulty. The whole study proposes a combination of spiked and normal convolution layers as an energy-efficient and reliable object detector model. The proposed model is split into two sections. The first section is developed as a feature extractor, which utilizes pre-trained VGG16, and the second section of the proposal structure is the combination of spiked and normal Convolutional layers to detect the bounding boxes of images. We drew a pre-trained model for classifying detected objects. With state of the art Python libraries, spike layers can be trained efficiently. The proposed spike convolutional object detector (SCOD) has been evaluated on VOC and Ex-Dark datasets. SCOD reached 66.01% and 41.25% mAP for detecting 20 different objects in the VOC-12 and 12 objects in the Ex-Dark dataset. SCOD uses 14 Giga FLOPS for its forward path calculations. Experimental results indicated superior performance compared to Tiny YOLO, Spike YOLO, YOLO-LITE, Tinier YOLO and Center of loc+Xception based on mAP for the VOC dataset.
Rethinking Model Selection and Decoding for Keyphrase Generation with Pre-trained Sequence-to-Sequence Models
Abstract
Keyphrase Generation (KPG) is a longstanding task in NLP with widespread applications. The advent of sequence-to-sequence (seq2seq) pre-trained language models (PLMs) has ushered in a transformative era for KPG, yielding promising performance improvements. However, many design decisions remain unexplored and are often made arbitrarily. This paper undertakes a systematic analysis of the influence of model selection and decoding strategies on PLM-based KPG. We begin by elucidating why seq2seq PLMs are apt for KPG, anchored by an attention-driven hypothesis. We then establish that conventional wisdom for selecting seq2seq PLMs lacks depth: (1) merely increasing model size or performing task-specific adaptation is not parameter-efficient; (2) although combining in-domain pre-training with task adaptation benefits KPG, it does partially hinder generalization. Regarding decoding, we demonstrate that while greedy search delivers strong F1 scores, it lags in recall compared with sampling-based methods. From our insights, we propose DeSel, a likelihood-based decode-select algorithm that improves greedy search by an average of 4.7% semantic F1 across five datasets. Our collective findings pave the way for deeper future investigations into PLM-based KPG.
What Makes for Robust Multi-Modal Models in the Face of Missing Modalities?
Authors: Siting Li, Chenzhuang Du, Yue Zhao, Yu Huang, Hang Zhao
Abstract
With the growing success of multi-modal learning, research on the robustness of multi-modal models, especially when facing situations with missing modalities, is receiving increased attention. Nevertheless, previous studies in this domain exhibit certain limitations, as they often lack theoretical insights or their methodologies are tied to specific network architectures or modalities. We model the scenarios of multi-modal models encountering missing modalities from an information-theoretic perspective and illustrate that the performance ceiling in such scenarios can be approached by efficiently utilizing the information inherent in non-missing modalities. In practice, there are two key aspects: (1) The encoder should be able to extract sufficiently good features from the non-missing modality; (2) The extracted features should be robust enough not to be influenced by noise during the fusion process across modalities. To this end, we introduce Uni-Modal Ensemble with Missing Modality Adaptation (UME-MMA). UME-MMA employs uni-modal pre-trained weights for the multi-modal model to enhance feature extraction and utilizes missing modality data augmentation techniques to better adapt to situations with missing modalities. Apart from that, UME-MMA, built on a late-fusion learning framework, allows for the plug-and-play use of various encoders, making it suitable for a wide range of modalities and enabling seamless integration of large-scale pre-trained encoders to further enhance performance. And we demonstrate UME-MMA's effectiveness in audio-visual datasets~(e.g., AV-MNIST, Kinetics-Sound, AVE) and vision-language datasets~(e.g., MM-IMDB, UPMC Food101).
Learning Stackable and Skippable LEGO Bricks for Efficient, Reconfigurable, and Variable-Resolution Diffusion Modeling
Abstract
Diffusion models excel at generating photo-realistic images but come with significant computational costs in both training and sampling. While various techniques address these computational challenges, a less-explored issue is designing an efficient and adaptable network backbone for iterative refinement. Current options like U-Net and Vision Transformer often rely on resource-intensive deep networks and lack the flexibility needed for generating images at variable resolutions or with a smaller network than used in training. This study introduces LEGO bricks, which seamlessly integrate Local-feature Enrichment and Global-content Orchestration. These bricks can be stacked to create a test-time reconfigurable diffusion backbone, allowing selective skipping of bricks to reduce sampling costs and generate higher-resolution images than the training data. LEGO bricks enrich local regions with an MLP and transform them using a Transformer block while maintaining a consistent full-resolution image across all bricks. Experimental results demonstrate that LEGO bricks enhance training efficiency, expedite convergence, and facilitate variable-resolution image generation while maintaining strong generative performance. Moreover, LEGO significantly reduces sampling time compared to other methods, establishing it as a valuable enhancement for diffusion models.
Top of the Heap: Efficient Memory Error Protection for Many Heap Objects
Authors: Kaiming Huang, Mathias Payer, Zhiyun Qian, Jack Sampson, Gang Tan, Trent Jaeger
Abstract
Exploits against heap memory errors continue to be a major concern. Although many defenses have been proposed, heap data are not protected from attacks that exploit memory errors systematically. Research defenses focus on complete coverage of heap objects, often giving up on comprehensive memory safety protection and/or incurring high costs in performance overhead and memory usage. In this paper, we propose a solution for heap memory safety enforcement that aims to provide comprehensive protection from memory errors efficiently by protecting those heap objects whose accesses are provably safe from memory errors. Specifically, we present the Uriah system that statically validates spatial and type memory safety for heap objects, isolating compliant objects on a safe heap that enforces temporal type safety to prevent attacks on memory reuse. Using Uriah, 71.9% of heap allocation sites can be shown to produce objects (73% of allocations are found safe) that satisfy spatial and type safety, which are then isolated using Uriah's heap allocator from memory accesses via unsafe heap objects. Uriah only incurs 2.9% overhead and only uses 9.3% more memory on SPEC CPU2006 (C/C++) benchmarks, showing that many heap objects can be protected from all classes of memory errors efficiently.
TANGO: Time-Reversal Latent GraphODE for Multi-Agent Dynamical Systems
Abstract
Learning complex multi-agent system dynamics from data is crucial across many domains, such as in physical simulations and material modeling. Extended from purely data-driven approaches, existing physics-informed approaches such as Hamiltonian Neural Network strictly follow energy conservation law to introduce inductive bias, making their learning more sample efficiently. However, many real-world systems do not strictly conserve energy, such as spring systems with frictions. Recognizing this, we turn our attention to a broader physical principle: Time-Reversal Symmetry, which depicts that the dynamics of a system shall remain invariant when traversed back over time. It still helps to preserve energies for conservative systems and in the meanwhile, serves as a strong inductive bias for non-conservative, reversible systems. To inject such inductive bias, in this paper, we propose a simple-yet-effective self-supervised regularization term as a soft constraint that aligns the forward and backward trajectories predicted by a continuous graph neural network-based ordinary differential equation (GraphODE). It effectively imposes time-reversal symmetry to enable more accurate model predictions across a wider range of dynamical systems under classical mechanics. In addition, we further provide theoretical analysis to show that our regularization essentially minimizes higher-order Taylor expansion terms during the ODE integration steps, which enables our model to be more noise-tolerant and even applicable to irreversible systems. Experimental results on a variety of physical systems demonstrate the effectiveness of our proposed method. Particularly, it achieves an MSE improvement of 11.5 % on a challenging chaotic triple-pendulum systems.
Whispering LLaMA: A Cross-Modal Generative Error Correction Framework for Speech Recognition
Authors: Srijith Radhakrishnan, Chao-Han Huck Yang, Sumeer Ahmad Khan, Rohit Kumar, Narsis A. Kiani, David Gomez-Cabrero, Jesper N. Tegner
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Multimedia (cs.MM); Sound (cs.SD); Audio and Speech Processing (eess.AS)
Abstract
We introduce a new cross-modal fusion technique designed for generative error correction in automatic speech recognition (ASR). Our methodology leverages both acoustic information and external linguistic representations to generate accurate speech transcription contexts. This marks a step towards a fresh paradigm in generative error correction within the realm of n-best hypotheses. Unlike the existing ranking-based rescoring methods, our approach adeptly uses distinct initialization techniques and parameter-efficient algorithms to boost ASR performance derived from pre-trained speech and text models. Through evaluation across diverse ASR datasets, we evaluate the stability and reproducibility of our fusion technique, demonstrating its improved word error rate relative (WERR) performance in comparison to n-best hypotheses by relatively 37.66%. To encourage future research, we have made our code and pre-trained models open source at https://github.com/Srijith-rkr/Whispering-LLaMA.
MemSum-DQA: Adapting An Efficient Long Document Extractive Summarizer for Document Question Answering
Authors: Nianlong Gu, Yingqiang Gao, Richard H. R. Hahnloser
Abstract
We introduce MemSum-DQA, an efficient system for document question answering (DQA) that leverages MemSum, a long document extractive summarizer. By prefixing each text block in the parsed document with the provided question and question type, MemSum-DQA selectively extracts text blocks as answers from documents. On full-document answering tasks, this approach yields a 9% improvement in exact match accuracy over prior state-of-the-art baselines. Notably, MemSum-DQA excels in addressing questions related to child-relationship understanding, underscoring the potential of extractive summarization techniques for DQA tasks.
Abstract
Machine learning models need to be continually updated or corrected to ensure that the prediction accuracy remains consistently high. In this study, we consider scenarios where developers should be careful to change the prediction results by the model correction, such as when the model is part of a complex system or software. In such scenarios, the developers want to control the specification of the corrections. To achieve this, the developers need to understand which subpopulations of the inputs get inaccurate predictions by the model. Therefore, we propose correction rule mining to acquire a comprehensive list of rules that describe inaccurate subpopulations and how to correct them. We also develop an efficient correction rule mining algorithm that is a combination of frequent itemset mining and a unique pruning technique for correction rules. We observed that the proposed algorithm found various rules which help to collect data insufficiently learned, directly correct model outputs, and analyze concept drift.
A Geometrical Approach to Evaluate the Adversarial Robustness of Deep Neural Networks
Authors: Yang Wang, Bo Dong, Ke Xu, Haiyin Piao, Yufei Ding, Baocai Yin, Xin Yang
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Deep Neural Networks (DNNs) are widely used for computer vision tasks. However, it has been shown that deep models are vulnerable to adversarial attacks, i.e., their performances drop when imperceptible perturbations are made to the original inputs, which may further degrade the following visual tasks or introduce new problems such as data and privacy security. Hence, metrics for evaluating the robustness of deep models against adversarial attacks are desired. However, previous metrics are mainly proposed for evaluating the adversarial robustness of shallow networks on the small-scale datasets. Although the Cross Lipschitz Extreme Value for nEtwork Robustness (CLEVER) metric has been proposed for large-scale datasets (e.g., the ImageNet dataset), it is computationally expensive and its performance relies on a tractable number of samples. In this paper, we propose the Adversarial Converging Time Score (ACTS), an attack-dependent metric that quantifies the adversarial robustness of a DNN on a specific input. Our key observation is that local neighborhoods on a DNN's output surface would have different shapes given different inputs. Hence, given different inputs, it requires different time for converging to an adversarial sample. Based on this geometry meaning, ACTS measures the converging time as an adversarial robustness metric. We validate the effectiveness and generalization of the proposed ACTS metric against different adversarial attacks on the large-scale ImageNet dataset using state-of-the-art deep networks. Extensive experiments show that our ACTS metric is an efficient and effective adversarial metric over the previous CLEVER metric.
Topological data analysis of human vowels: Persistent homologies across representation spaces
Authors: Guillem Bonafos, Jean-Marc Freyermuth, Pierre Pudlo, Samuel Tronçon, Arnaud Rey
Abstract
Topological Data Analysis (TDA) has been successfully used for various tasks in signal/image processing, from visualization to supervised/unsupervised classification. Often, topological characteristics are obtained from persistent homology theory. The standard TDA pipeline starts from the raw signal data or a representation of it. Then, it consists in building a multiscale topological structure on the top of the data using a pre-specified filtration, and finally to compute the topological signature to be further exploited. The commonly used topological signature is a persistent diagram (or transformations of it). Current research discusses the consequences of the many ways to exploit topological signatures, much less often the choice of the filtration, but to the best of our knowledge, the choice of the representation of a signal has not been the subject of any study yet. This paper attempts to provide some answers on the latter problem. To this end, we collected real audio data and built a comparative study to assess the quality of the discriminant information of the topological signatures extracted from three different representation spaces. Each audio signal is represented as i) an embedding of observed data in a higher dimensional space using Taken's representation, ii) a spectrogram viewed as a surface in a 3D ambient space, iii) the set of spectrogram's zeroes. From vowel audio recordings, we use topological signature for three prediction problems: speaker gender, vowel type, and individual. We show that topologically-augmented random forest improves the Out-of-Bag Error (OOB) over solely based Mel-Frequency Cepstral Coefficients (MFCC) for the last two problems. Our results also suggest that the topological information extracted from different signal representations is complementary, and that spectrogram's zeros offers the best improvement for gender prediction.
Self-Supervised Set Representation Learning for Unsupervised Meta-Learning
Authors: Dong Bok Lee, Seanie Lee, Joonho Ko, Kenji Kawaguchi, Juho Lee, Sung Ju Hwang
Abstract
Dataset distillation methods have achieved remarkable success in distilling a large dataset into a small set of representative samples. However, they are not designed to produce a distilled dataset that can be effectively used for facilitating self-supervised pre-training. To this end, we propose a novel problem of distilling an unlabeled dataset into a set of small synthetic samples for efficient self-supervised learning (SSL). We first prove that a gradient of synthetic samples with respect to a SSL objective in naive bilevel optimization is \textit{biased} due to the randomness originating from data augmentations or masking. To address this issue, we propose to minimize the mean squared error (MSE) between a model's representations of the synthetic examples and their corresponding learnable target feature representations for the inner objective, which does not introduce any randomness. Our primary motivation is that the model obtained by the proposed inner optimization can mimic the \textit{self-supervised target model}. To achieve this, we also introduce the MSE between representations of the inner model and the self-supervised target model on the original full dataset for outer optimization. Lastly, assuming that a feature extractor is fixed, we only optimize a linear head on top of the feature extractor, which allows us to reduce the computational cost and obtain a closed-form solution of the head with kernel ridge regression. We empirically validate the effectiveness of our method on various applications involving transfer learning.
Watt For What: Rethinking Deep Learning's Energy-Performance Relationship
Authors: Shreyank N Gowda, Xinyue Hao, Gen Li, Laura Sevilla-Lara, Shashank Narayana Gowda
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Deep learning models have revolutionized various fields, from image recognition to natural language processing, by achieving unprecedented levels of accuracy. However, their increasing energy consumption has raised concerns about their environmental impact, disadvantaging smaller entities in research and exacerbating global energy consumption. In this paper, we explore the trade-off between model accuracy and electricity consumption, proposing a metric that penalizes large consumption of electricity. We conduct a comprehensive study on the electricity consumption of various deep learning models across different GPUs, presenting a detailed analysis of their accuracy-efficiency trade-offs. By evaluating accuracy per unit of electricity consumed, we demonstrate how smaller, more energy-efficient models can significantly expedite research while mitigating environmental concerns. Our results highlight the potential for a more sustainable approach to deep learning, emphasizing the importance of optimizing models for efficiency. This research also contributes to a more equitable research landscape, where smaller entities can compete effectively with larger counterparts. This advocates for the adoption of efficient deep learning practices to reduce electricity consumption, safeguarding the environment for future generations whilst also helping ensure a fairer competitive landscape.
Realizing Stabilized Landing for Computation-Limited Reusable Rockets: A Quantum Reinforcement Learning Approach
Authors: Gyu Seon Kim, JaeHyun Chung, Soohyun Park
Abstract
The advent of reusable rockets has heralded a new era in space exploration, reducing the costs of launching satellites by a significant factor. Traditional rockets were disposable, but the design of reusable rockets for repeated use has revolutionized the financial dynamics of space missions. The most critical phase of reusable rockets is the landing stage, which involves managing the tremendous speed and attitude for safe recovery. The complexity of this task presents new challenges for control systems, specifically in terms of precision and adaptability. Classical control systems like the proportional-integral-derivative (PID) controller lack the flexibility to adapt to dynamic system changes, making them costly and time-consuming to redesign of controller. This paper explores the integration of quantum reinforcement learning into the control systems of reusable rockets as a promising alternative. Unlike classical reinforcement learning, quantum reinforcement learning uses quantum bits that can exist in superposition, allowing for more efficient information encoding and reducing the number of parameters required. This leads to increased computational efficiency, reduced memory requirements, and more stable and predictable performance. Due to the nature of reusable rockets, which must be light, heavy computers cannot fit into them. In the reusable rocket scenario, quantum reinforcement learning, which has reduced memory requirements due to fewer parameters, is a good solution.
Efficient Retrieval of Images with Irregular Patterns using Morphological Image Analysis: Applications to Industrial and Healthcare datasets
Authors: Jiajun Zhang, Georgina Cosma, Sarah Bugby, Jason Watkins
Subjects: Information Retrieval (cs.IR); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Image retrieval is the process of searching and retrieving images from a database based on their visual content and features. Recently, much attention has been directed towards the retrieval of irregular patterns within industrial or medical images by extracting features from the images, such as deep features, colour-based features, shape-based features and local features. This has applications across a spectrum of industries, including fault inspection, disease diagnosis, and maintenance prediction. This paper proposes an image retrieval framework to search for images containing similar irregular patterns by extracting a set of morphological features (DefChars) from images; the datasets employed in this paper contain wind turbine blade images with defects, chest computerised tomography scans with COVID-19 infection, heatsink images with defects, and lake ice images. The proposed framework was evaluated with different feature extraction methods (DefChars, resized raw image, local binary pattern, and scale-invariant feature transforms) and distance metrics to determine the most efficient parameters in terms of retrieval performance across datasets. The retrieval results show that the proposed framework using the DefChars and the Manhattan distance metric achieves a mean average precision of 80% and a low standard deviation of 0.09 across classes of irregular patterns, outperforming alternative feature-metric combinations across all datasets. Furthermore, the low standard deviation between each class highlights DefChars' capability for a reliable image retrieval task, even in the presence of class imbalances or small-sized datasets.
Energy-Efficient Visual Search by Eye Movement and Low-Latency Spiking Neural Network
Authors: Yunhui Zhou, Dongqi Han, Yuguo Yu
Subjects: Neural and Evolutionary Computing (cs.NE); Computer Vision and Pattern Recognition (cs.CV); Neurons and Cognition (q-bio.NC)
Abstract
Human vision incorporates non-uniform resolution retina, efficient eye movement strategy, and spiking neural network (SNN) to balance the requirements in visual field size, visual resolution, energy cost, and inference latency. These properties have inspired interest in developing human-like computer vision. However, existing models haven't fully incorporated the three features of human vision, and their learned eye movement strategies haven't been compared with human's strategy, making the models' behavior difficult to interpret. Here, we carry out experiments to examine human visual search behaviors and establish the first SNN-based visual search model. The model combines an artificial retina with spiking feature extraction, memory, and saccade decision modules, and it employs population coding for fast and efficient saccade decisions. The model can learn either a human-like or a near-optimal fixation strategy, outperform humans in search speed and accuracy, and achieve high energy efficiency through short saccade decision latency and sparse activation. It also suggests that the human search strategy is suboptimal in terms of search speed. Our work connects modeling of vision in neuroscience and machine learning and sheds light on developing more energy-efficient computer vision algorithms.
FTFT: efficient and robust Fine-Tuning by transFerring Training dynamics
Authors: Yupei Du, Albert Gatt, Dong Nguyen
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Abstract
Despite the massive success of fine-tuning large Pre-trained Language Models (PLMs) on a wide range of Natural Language Processing (NLP) tasks, they remain susceptible to out-of-distribution (OOD) and adversarial inputs. Data map (DM) is a simple yet effective dual-model approach that enhances the robustness of fine-tuned PLMs, which involves fine-tuning a model on the original training set (i.e. reference model), selecting a specified fraction of important training examples according to the training dynamics of the reference model, and fine-tuning the same model on these selected examples (i.e. main model). However, it suffers from the drawback of requiring fine-tuning the same model twice, which is computationally expensive for large models. In this paper, we first show that 1) training dynamics are highly transferable across different model sizes and different pre-training methods, and that 2) main models fine-tuned using DM learn faster than when using conventional Empirical Risk Minimization (ERM). Building on these observations, we propose a novel fine-tuning approach based on the DM method: Fine-Tuning by transFerring Training dynamics (FTFT). Compared with DM, FTFT uses more efficient reference models and then fine-tunes more capable main models for fewer steps. Our experiments show that FTFT achieves better generalization robustness than ERM while spending less than half of the training cost.
EViT: An Eagle Vision Transformer with Bi-Fovea Self-Attention
Abstract
Because of the advancement of deep learning technology, vision transformer has demonstrated competitive performance in various computer vision tasks. Unfortunately, vision transformer still faces some challenges such as high computational complexity and absence of desirable inductive bias. To alleviate these problems, this study proposes a novel Bi-Fovea Self-Attention (BFSA) inspired by the physiological structure and characteristics of bi-fovea vision in eagle eyes. This BFSA can simulate the shallow fovea and deep fovea functions of eagle vision, enabling the network to extract feature representations of targets from coarse to fine, facilitating the interaction of multi-scale feature representations. Additionally, this study designs a Bionic Eagle Vision (BEV) block based on BFSA and CNN. It combines CNN and Vision Transformer, to enhance the network's local and global representation ability for targets. Furthermore, this study develops a unified and efficient general pyramid backbone network family, named Eagle Vision Transformers (EViTs) by stacking the BEV blocks. Experimental results on various computer vision tasks including image classification, object detection, instance segmentation and other transfer learning tasks show that the proposed EViTs perform significantly better than the baselines under similar model sizes, which exhibits faster speed on graphics processing unit compared to other models. Code will be released at https://github.com/nkusyl.
A Survey on Intelligent Iterative Methods for Solving Sparse Linear Algebraic Equations
Abstract
Efficiently solving sparse linear algebraic equations is an important research topic of numerical simulation. Commonly used approaches include direct methods and iterative methods. Compared with the direct methods, the iterative methods have lower computational complexity and memory consumption, and are thus often used to solve large-scale sparse linear equations. However, there are numerous iterative methods, parameters and components needed to be carefully chosen, and an inappropriate combination may eventually lead to an inefficient solution process in practice. With the development of deep learning, intelligent iterative methods become popular in these years, which can intelligently make a sufficiently good combination, optimize the parameters and components in accordance with the properties of the input matrix. This survey then reviews these intelligent iterative methods. To be clearer, we shall divide our discussion into three aspects: a method aspect, a component aspect and a parameter aspect. Moreover, we summarize the existing work and propose potential research directions that may deserve a deep investigation.
The Lattice Overparametrization Paradigm for the Machine Learning of Lattice Operators
Abstract
The machine learning of lattice operators has three possible bottlenecks. From a statistical standpoint, it is necessary to design a constrained class of operators based on prior information with low bias, and low complexity relative to the sample size. From a computational perspective, there should be an efficient algorithm to minimize an empirical error over the class. From an understanding point of view, the properties of the learned operator need to be derived, so its behavior can be theoretically understood. The statistical bottleneck can be overcome due to the rich literature about the representation of lattice operators, but there is no general learning algorithm for them. In this paper, we discuss a learning paradigm in which, by overparametrizing a class via elements in a lattice, an algorithm for minimizing functions in a lattice is applied to learn. We present the stochastic lattice gradient descent algorithm as a general algorithm to learn on constrained classes of operators as long as a lattice overparametrization of it is fixed, and we discuss previous works which are proves of concept. Moreover, if there are algorithms to compute the basis of an operator from its overparametrization, then its properties can be deduced and the understanding bottleneck is also overcome. This learning paradigm has three properties that modern methods based on neural networks lack: control, transparency and interpretability. Nowadays, there is an increasing demand for methods with these characteristics, and we believe that mathematical morphology is in a unique position to supply them. The lattice overparametrization paradigm could be a missing piece for it to achieve its full potential within modern machine learning.
Self-Supervised Representation Learning for Online Handwriting Text Classification
Abstract
Self-supervised learning offers an efficient way of extracting rich representations from various types of unlabeled data while avoiding the cost of annotating large-scale datasets. This is achievable by designing a pretext task to form pseudo labels with respect to the modality and domain of the data. Given the evolving applications of online handwritten texts, in this study, we propose the novel Part of Stroke Masking (POSM) as a pretext task for pretraining models to extract informative representations from the online handwriting of individuals in English and Chinese languages, along with two suggested pipelines for fine-tuning the pretrained models. To evaluate the quality of the extracted representations, we use both intrinsic and extrinsic evaluation methods. The pretrained models are fine-tuned to achieve state-of-the-art results in tasks such as writer identification, gender classification, and handedness classification, also highlighting the superiority of utilizing the pretrained models over the models trained from scratch.
A Parallelized, Adam-Based Solver for Reserve and Security Constrained AC Unit Commitment
Abstract
Power system optimization problems which include the nonlinear AC power flow equations require powerful and robust numerical solution algorithms. Within this sub-field of nonlinear optimization, interior point methods have come to dominate the solver landscape. Over the last decade, however, a number of efficient numerical optimizers have emerged from the field of Machine Learning (ML). One algorithm in particular, Adam, has become the optimizer-of-choice for a massive percentage of ML training problems (including, e.g., the training of GPT-3), solving some of the largest unconstrained optimization problems ever conceived of. Inspired by such progress, this paper designs a parallelized Adam-based numerical solver to overcome one of the most challenging power system optimization problems: security and reserve constrained AC Unit Commitment. The resulting solver, termed quasiGrad, recently competed in the third ARPA-E Grid Optimization (GO3) competition. In the day-ahead market clearing category (with systems ranging from 3 to 23,643 buses over 48 time periods), quasiGrad's aggregated market surplus scores were within 5% of the winningest market surplus scores. The quasiGrad solver is now released as an open-source Julia package: quasiGrad.jl. The internal gradient-based solver (Adam) can easily be substituted for other ML-inspired solvers (e.g., AdaGrad, AdaDelta, RMSProp, etc.). Test results from large experiments are provided.
Unlock the Potential of Counterfactually-Augmented Data in Out-Of-Distribution Generalization
Abstract
Counterfactually-Augmented Data (CAD) -- minimal editing of sentences to flip the corresponding labels -- has the potential to improve the Out-Of-Distribution (OOD) generalization capability of language models, as CAD induces language models to exploit domain-independent causal features and exclude spurious correlations. However, the empirical results of CAD's OOD generalization are not as efficient as anticipated. In this study, we attribute the inefficiency to the myopia phenomenon caused by CAD: language models only focus on causal features that are edited in the augmentation operation and exclude other non-edited causal features. Therefore, the potential of CAD is not fully exploited. To address this issue, we analyze the myopia phenomenon in feature space from the perspective of Fisher's Linear Discriminant, then we introduce two additional constraints based on CAD's structural properties (dataset-level and sentence-level) to help language models extract more complete causal features in CAD, thereby mitigating the myopia phenomenon and improving OOD generalization capability. We evaluate our method on two tasks: Sentiment Analysis and Natural Language Inference, and the experimental results demonstrate that our method could unlock the potential of CAD and improve the OOD generalization performance of language models by 1.0% to 5.9%.
Machine Learning Quantum Systems with Magnetic p-bits
Abstract
The slowing down of Moore's Law has led to a crisis as the computing workloads of Artificial Intelligence (AI) algorithms continue skyrocketing. There is an urgent need for scalable and energy-efficient hardware catering to the unique requirements of AI algorithms and applications. In this environment, probabilistic computing with p-bits emerged as a scalable, domain-specific, and energy-efficient computing paradigm, particularly useful for probabilistic applications and algorithms. In particular, spintronic devices such as stochastic magnetic tunnel junctions (sMTJ) show great promise in designing integrated p-computers. Here, we examine how a scalable probabilistic computer with such magnetic p-bits can be useful for an emerging field combining machine learning and quantum physics.
DeepLSH: Deep Locality-Sensitive Hash Learning for Fast and Efficient Near-Duplicate Crash Report Detection
Abstract
Automatic crash bucketing is a crucial phase in the software development process for efficiently triaging bug reports. It generally consists in grouping similar reports through clustering techniques. However, with real-time streaming bug collection, systems are needed to quickly answer the question: What are the most similar bugs to a new one?, that is, efficiently find near-duplicates. It is thus natural to consider nearest neighbors search to tackle this problem and especially the well-known locality-sensitive hashing (LSH) to deal with large datasets due to its sublinear performance and theoretical guarantees on the similarity search accuracy. Surprisingly, LSH has not been considered in the crash bucketing literature. It is indeed not trivial to derive hash functions that satisfy the so-called locality-sensitive property for the most advanced crash bucketing metrics. Consequently, we study in this paper how to leverage LSH for this task. To be able to consider the most relevant metrics used in the literature, we introduce DeepLSH, a Siamese DNN architecture with an original loss function, that perfectly approximates the locality-sensitivity property even for Jaccard and Cosine metrics for which exact LSH solutions exist. We support this claim with a series of experiments on an original dataset, which we make available.
Comparing AI Algorithms for Optimizing Elliptic Curve Cryptography Parameters in Third-Party E-Commerce Integrations: A Pre-Quantum Era Analysis
Authors: Felipe Tellez, Jorge Ortiz
Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
Abstract
This paper presents a comparative analysis between the Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), two vital artificial intelligence algorithms, focusing on optimizing Elliptic Curve Cryptography (ECC) parameters. These encompass the elliptic curve coefficients, prime number, generator point, group order, and cofactor. The study provides insights into which of the bio-inspired algorithms yields better optimization results for ECC configurations, examining performances under the same fitness function. This function incorporates methods to ensure robust ECC parameters, including assessing for singular or anomalous curves and applying Pollard's rho attack and Hasse's theorem for optimization precision. The optimized parameters generated by GA and PSO are tested in a simulated e-commerce environment, contrasting with well-known curves like secp256k1 during the transmission of order messages using Elliptic Curve-Diffie Hellman (ECDH) and Hash-based Message Authentication Code (HMAC). Focusing on traditional computing in the pre-quantum era, this research highlights the efficacy of GA and PSO in ECC optimization, with implications for enhancing cybersecurity in third-party e-commerce integrations. We recommend the immediate consideration of these findings before quantum computing's widespread adoption.
Going Beyond Neural Network Feature Similarity: The Network Feature Complexity and Its Interpretation Using Category Theory
Abstract
The behavior of neural networks still remains opaque, and a recently widely noted phenomenon is that networks often achieve similar performance when initialized with different random parameters. This phenomenon has attracted significant attention in measuring the similarity between features learned by distinct networks. However, feature similarity could be vague in describing the same feature since equivalent features hardly exist. In this paper, we expand the concept of equivalent feature and provide the definition of what we call functionally equivalent features. These features produce equivalent output under certain transformations. Using this definition, we aim to derive a more intrinsic metric for the so-called feature complexity regarding the redundancy of features learned by a neural network at each layer. We offer a formal interpretation of our approach through the lens of category theory, a well-developed area in mathematics. To quantify the feature complexity, we further propose an efficient algorithm named Iterative Feature Merging. Our experimental results validate our ideas and theories from various perspectives. We empirically demonstrate that the functionally equivalence widely exists among different features learned by the same neural network and we could reduce the number of parameters of the network without affecting the performance.The IFM shows great potential as a data-agnostic model prune method. We have also drawn several interesting empirical findings regarding the defined feature complexity.
Correlated Noise Provably Beats Independent Noise for Differentially Private Learning
Authors: Christopher A. Choquette-Choo, Krishnamurthy Dvijotham, Krishna Pillutla, Arun Ganesh, Thomas Steinke, Abhradeep Thakurta
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Optimization and Control (math.OC)
Abstract
Differentially private learning algorithms inject noise into the learning process. While the most common private learning algorithm, DP-SGD, adds independent Gaussian noise in each iteration, recent work on matrix factorization mechanisms has shown empirically that introducing correlations in the noise can greatly improve their utility. We characterize the asymptotic learning utility for any choice of the correlation function, giving precise analytical bounds for linear regression and as the solution to a convex program for general convex functions. We show, using these bounds, how correlated noise provably improves upon vanilla DP-SGD as a function of problem parameters such as the effective dimension and condition number. Moreover, our analytical expression for the near-optimal correlation function circumvents the cubic complexity of the semi-definite program used to optimize the noise correlation matrix in previous work. We validate our theory with experiments on private deep learning. Our work matches or outperforms prior work while being efficient both in terms of compute and memory.
Uni3D: Exploring Unified 3D Representation at Scale
Abstract
Scaling up representations for images or text has been extensively investigated in the past few years and has led to revolutions in learning vision and language. However, scalable representation for 3D objects and scenes is relatively unexplored. In this work, we present Uni3D, a 3D foundation model to explore the unified 3D representation at scale. Uni3D uses a 2D initialized ViT end-to-end pretrained to align the 3D point cloud features with the image-text aligned features. Via the simple architecture and pretext task, Uni3D can leverage abundant 2D pretrained models as initialization and image-text aligned models as the target, unlocking the great potential of 2D models and scaling-up strategies to the 3D world. We efficiently scale up Uni3D to one billion parameters, and set new records on a broad range of 3D tasks, such as zero-shot classification, few-shot classification, open-world understanding and part segmentation. We show that the strong Uni3D representation also enables applications such as 3D painting and retrieval in the wild. We believe that Uni3D provides a new direction for exploring both scaling up and efficiency of the representation in 3D domain.
Information Content Exploration
Authors: Jacob Chmura, Hasham Burhani, Xiao Qi Shi
Abstract
Sparse reward environments are known to be challenging for reinforcement learning agents. In such environments, efficient and scalable exploration is crucial. Exploration is a means by which an agent gains information about the environment. We expand on this topic and propose a new intrinsic reward that systemically quantifies exploratory behavior and promotes state coverage by maximizing the information content of a trajectory taken by an agent. We compare our method to alternative exploration based intrinsic reward techniques, namely Curiosity Driven Learning and Random Network Distillation. We show that our information theoretic reward induces efficient exploration and outperforms in various games, including Montezuma Revenge, a known difficult task for reinforcement learning. Finally, we propose an extension that maximizes information content in a discretely compressed latent space which boosts sample efficiency and generalizes to continuous state spaces.
A Supervised Embedding and Clustering Anomaly Detection method for classification of Mobile Network Faults
Abstract
The paper introduces Supervised Embedding and Clustering Anomaly Detection (SEMC-AD), a method designed to efficiently identify faulty alarm logs in a mobile network and alleviate the challenges of manual monitoring caused by the growing volume of alarm logs. SEMC-AD employs a supervised embedding approach based on deep neural networks, utilizing historical alarm logs and their labels to extract numerical representations for each log, effectively addressing the issue of imbalanced classification due to a small proportion of anomalies in the dataset without employing one-hot encoding. The robustness of the embedding is evaluated by plotting the two most significant principle components of the embedded alarm logs, revealing that anomalies form distinct clusters with similar embeddings. Multivariate normal Gaussian clustering is then applied to these components, identifying clusters with a high ratio of anomalies to normal alarms (above 90%) and labeling them as the anomaly group. To classify new alarm logs, we check if their embedded vectors' two most significant principle components fall within the anomaly-labeled clusters. If so, the log is classified as an anomaly. Performance evaluation demonstrates that SEMC-AD outperforms conventional random forest and gradient boosting methods without embedding. SEMC-AD achieves 99% anomaly detection, whereas random forest and XGBoost only detect 86% and 81% of anomalies, respectively. While supervised classification methods may excel in labeled datasets, the results demonstrate that SEMC-AD is more efficient in classifying anomalies in datasets with numerous categorical features, significantly enhancing anomaly detection, reducing operator burden, and improving network maintenance.
Spectral Entry-wise Matrix Estimation for Low-Rank Reinforcement Learning
Authors: Stefan Stojanovic, Yassir Jedra, Alexandre Proutiere
Abstract
We study matrix estimation problems arising in reinforcement learning (RL) with low-rank structure. In low-rank bandits, the matrix to be recovered specifies the expected arm rewards, and for low-rank Markov Decision Processes (MDPs), it may for example characterize the transition kernel of the MDP. In both cases, each entry of the matrix carries important information, and we seek estimation methods with low entry-wise error. Importantly, these methods further need to accommodate for inherent correlations in the available data (e.g. for MDPs, the data consists of system trajectories). We investigate the performance of simple spectral-based matrix estimation approaches: we show that they efficiently recover the singular subspaces of the matrix and exhibit nearly-minimal entry-wise error. These new results on low-rank matrix estimation make it possible to devise reinforcement learning algorithms that fully exploit the underlying low-rank structure. We provide two examples of such algorithms: a regret minimization algorithm for low-rank bandit problems, and a best policy identification algorithm for reward-free RL in low-rank MDPs. Both algorithms yield state-of-the-art performance guarantees.
$f$-Policy Gradients: A General Framework for Goal Conditioned RL using $f$-Divergences
Authors: Siddhant Agarwal, Ishan Durugkar, Peter Stone, Amy Zhang
Abstract
Goal-Conditioned Reinforcement Learning (RL) problems often have access to sparse rewards where the agent receives a reward signal only when it has achieved the goal, making policy optimization a difficult problem. Several works augment this sparse reward with a learned dense reward function, but this can lead to sub-optimal policies if the reward is misaligned. Moreover, recent works have demonstrated that effective shaping rewards for a particular problem can depend on the underlying learning algorithm. This paper introduces a novel way to encourage exploration called $f$-Policy Gradients, or $f$-PG. $f$-PG minimizes the f-divergence between the agent's state visitation distribution and the goal, which we show can lead to an optimal policy. We derive gradients for various f-divergences to optimize this objective. Our learning paradigm provides dense learning signals for exploration in sparse reward settings. We further introduce an entropy-regularized policy optimization objective, that we call $state$-MaxEnt RL (or $s$-MaxEnt RL) as a special case of our objective. We show that several metric-based shaping rewards like L2 can be used with $s$-MaxEnt RL, providing a common ground to study such metric-based shaping rewards with efficient exploration. We find that $f$-PG has better performance compared to standard policy gradient methods on a challenging gridworld as well as the Point Maze and FetchReach environments. More information on our website https://agarwalsiddhant10.github.io/projects/fpg.html.
Inverse Factorized Q-Learning for Cooperative Multi-agent Imitation Learning
Authors: Authors: The Viet Bui, Tien Mai, Thanh Hong Nguyen
Subjects: Machine Learning (cs.LG); Multiagent Systems (cs.MA)
Abstract
This paper concerns imitation learning (IL) (i.e, the problem of learning to mimic expert behaviors from demonstrations) in cooperative multi-agent systems. The learning problem under consideration poses several challenges, characterized by high-dimensional state and action spaces and intricate inter-agent dependencies. In a single-agent setting, IL has proven to be done efficiently through an inverse soft-Q learning process given expert demonstrations. However, extending this framework to a multi-agent context introduces the need to simultaneously learn both local value functions to capture local observations and individual actions, and a joint value function for exploiting centralized learning. In this work, we introduce a novel multi-agent IL algorithm designed to address these challenges. Our approach enables the centralized learning by leveraging mixing networks to aggregate decentralized Q functions. A main advantage of this approach is that the weights of the mixing networks can be trained using information derived from global states. We further establish conditions for the mixing networks under which the multi-agent objective function exhibits convexity within the Q function space. We present extensive experiments conducted on some challenging competitive and cooperative multi-agent game environments, including an advanced version of the Star-Craft multi-agent challenge (i.e., SMACv2), which demonstrates the effectiveness of our proposed algorithm compared to existing state-of-the-art multi-agent IL algorithms.
Teaching Language Models to Hallucinate Less with Synthetic Tasks
Authors: Erik Jones, Hamid Palangi, Clarisse Simões, Varun Chandrasekaran, Subhabrata Mukherjee, Arindam Mitra, Ahmed Awadallah, Ece Kamar
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Abstract
Large language models (LLMs) frequently hallucinate on abstractive summarization tasks such as document-based question-answering, meeting summarization, and clinical report generation, even though all necessary information is included in context. However, optimizing LLMs to hallucinate less on these tasks is challenging, as hallucination is hard to efficiently evaluate at each optimization step. In this work, we show that reducing hallucination on a synthetic task can also reduce hallucination on real-world downstream tasks. Our method, SynTra, first designs a synthetic task where hallucinations are easy to elicit and measure. It next optimizes the LLM's system message via prefix-tuning on the synthetic task, and finally transfers the system message to realistic, hard-to-optimize tasks. Across three realistic abstractive summarization tasks, SynTra reduces hallucination for two 13B-parameter LLMs using only a synthetic retrieval task for supervision. We also find that optimizing the system message rather than the model weights can be critical; fine-tuning the entire model on the synthetic task can counterintuitively increase hallucination. Overall, SynTra demonstrates that the extra flexibility of working with synthetic data can help mitigate undesired behaviors in practice.
Keyword: faster
Analysis of Learned Features and Framework for Potato Disease Detection
Abstract
For applications like plant disease detection, usually, a model is trained on publicly available data and tested on field data. This means that the test data distribution is not the same as the training data distribution, which affects the classifier performance adversely. We handle this dataset shift by ensuring that the features are learned from disease spots in the leaf or healthy regions, as applicable. This is achieved using a faster Region-based convolutional neural network (RCNN) as one of the solutions and an attention-based network as the other. The average classification accuracies of these classifiers are approximately 95% while evaluated on the test set corresponding to their training dataset. These classifiers also performed equivalently, with an average score of 84% on a dataset not seen during the training phase.
DockGame: Cooperative Games for Multimeric Rigid Protein Docking
Authors: Vignesh Ram Somnath, Pier Giuseppe Sessa, Maria Rodriguez Martinez, Andreas Krause
Abstract
Protein interactions and assembly formation are fundamental to most biological processes. Predicting the assembly structure from constituent proteins -- referred to as the protein docking task -- is thus a crucial step in protein design applications. Most traditional and deep learning methods for docking have focused mainly on binary docking, following either a search-based, regression-based, or generative modeling paradigm. In this paper, we focus on the less-studied multimeric (i.e., two or more proteins) docking problem. We introduce DockGame, a novel game-theoretic framework for docking -- we view protein docking as a cooperative game between proteins, where the final assembly structure(s) constitute stable equilibria w.r.t. the underlying game potential. Since we do not have access to the true potential, we consider two approaches - i) learning a surrogate game potential guided by physics-based energy functions and computing equilibria by simultaneous gradient updates, and ii) sampling from the Gibbs distribution of the true potential by learning a diffusion generative model over the action spaces (rotations and translations) of all proteins. Empirically, on the Docking Benchmark 5.5 (DB5.5) dataset, DockGame has much faster runtimes than traditional docking methods, can generate multiple plausible assembly structures, and achieves comparable performance to existing binary docking baselines, despite solving the harder task of coordinating multiple protein chains.
Words into Action: Learning Diverse Humanoid Robot Behaviors using Language Guided Iterative Motion Refinement
Abstract
Humanoid robots are well suited for human habitats due to their morphological similarity, but developing controllers for them is a challenging task that involves multiple sub-problems, such as control, planning and perception. In this paper, we introduce a method to simplify controller design by enabling users to train and fine-tune robot control policies using natural language commands. We first learn a neural network policy that generates behaviors given a natural language command, such as "walk forward", by combining Large Language Models (LLMs), motion retargeting, and motion imitation. Based on the synthesized motion, we iteratively fine-tune by updating the text prompt and querying LLMs to find the best checkpoint associated with the closest motion in history. We validate our approach using a simulated Digit humanoid robot and demonstrate learning of diverse motions, such as walking, hopping, and kicking, without the burden of complex reward engineering. In addition, we show that our iterative refinement enables us to learn 3x times faster than a naive formulation that learns from scratch.
Abstract
Power side-channel attacks exploit the dynamic power consumption of cryptographic operations to leak sensitive information of encryption hardware. Therefore, it is necessary to conduct power side-channel analysis for assessing the susceptibility of cryptographic systems and mitigating potential risks. Existing power side-channel analysis primarily focuses on post-silicon implementations, which are inflexible in addressing design flaws, leading to costly and time-consuming post-fabrication design re-spins. Hence, pre-silicon power side-channel analysis is required for early detection of vulnerabilities to improve design robustness. In this paper, we introduce SCAR, a novel pre-silicon power side-channel analysis framework based on Graph Neural Networks (GNN). SCAR converts register-transfer level (RTL) designs of encryption hardware into control-data flow graphs and use that to detect the design modules susceptible to side-channel leakage. Furthermore, we incorporate a deep learning-based explainer in SCAR to generate quantifiable and human-accessible explanation of our detection and localization decisions. We have also developed a fortification component as a part of SCAR that uses large-language models (LLM) to automatically generate and insert additional design code at the localized zone to shore up the side-channel leakage. When evaluated on popular encryption algorithms like AES, RSA, and PRESENT, and postquantum cryptography algorithms like Saber and CRYSTALS-Kyber, SCAR, achieves up to 94.49% localization accuracy, 100% precision, and 90.48% recall. Additionally, through explainability analysis, SCAR reduces features for GNN model training by 57% while maintaining comparable accuracy. We believe that SCAR will transform the security-critical hardware design cycle, resulting in faster design closure at a reduced design cost.
Exploring Users Pointing Performance on Large Displays with Different Curvatures in Virtual Reality
Authors: A K M Amanat Ullah, William Delamare, Khalad Hasan
Abstract
Large curved displays inside Virtual Reality environments are becoming popular for visualizing high-resolution content during analytical tasks, gaming or entertainment. Prior research showed that such displays provide a wide field of view and offer users a high level of immersion. However, little is known about users' performance (e.g., pointing speed and accuracy) on them. We explore users' pointing performance on large virtual curved displays. We investigate standard pointing factors (e.g., target width and amplitude) in combination with relevant curve-related factors, namely display curvature and both linear and angular measures. Our results show that the less curved the display, the higher the performance, i.e., faster movement time. This result holds for pointing tasks controlled via their visual properties (linear widths and amplitudes) or their motor properties (angular widths and amplitudes). Additionally, display curvatures significantly affect the error rate for both linear and angular conditions. Furthermore, we observe that curved displays perform better or similar to flat displays based on throughput analysis. Finally, we discuss our results and provide suggestions regarding pointing tasks on large curved displays in VR.
Just-in-Time Flaky Test Detection via Abstracted Failure Symptom Matching
Authors: Gabin An, Juyeon Yoon, Thomas Bach, Jingun Hong, Shin Yoo
Abstract
We report our experience of using failure symptoms, such as error messages or stack traces, to identify flaky test failures in a Continuous Integration (CI) pipeline for a large industrial software system, SAP HANA. Although failure symptoms are commonly used to identify similar failures, they have not previously been employed to detect flaky test failures. Our hypothesis is that flaky failures will exhibit symptoms distinct from those of non-flaky failures. Consequently, we can identify recurring flaky failures, without rerunning the tests, by matching the failure symptoms to those of historical flaky runs. This can significantly reduce the need for test reruns, ultimately resulting in faster delivery of test results to developers. To facilitate the process of matching flaky failures across different execution instances, we abstract newer test failure symptoms before matching them to the known patterns of flaky failures, inspired by previous research in the fields of failure deduplication and log analysis. We evaluate our symptom-based flakiness detection method using actual failure symptoms gathered from CI data of SAP HANA during a six-month period. Our method shows the potential of using failure symptoms to identify recurring flaky failures, achieving a precision of at least 96%, while saving approximately 58% of the machine time compared to the traditional rerun strategy. Analysis of the false positives and the feedback from developers underscore the importance of having descriptive and informative failure symptoms for both the effective deployment of this symptom-based approach and the debugging of flaky tests.
Deep Learning for Automatic Detection and Facial Recognition in Japanese Macaques: Illuminating Social Networks
Authors: Julien Paulet (UJM), Axel Molina (ENS-PSL), Benjamin Beltzung (IPHC), Takafumi Suzumura, Shinya Yamamoto, Cédric Sueur (IPHC, IUF, ANTHROPO LAB)
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Social and Information Networks (cs.SI)
Abstract
Individual identification plays a pivotal role in ecology and ethology, notably as a tool for complex social structures understanding. However, traditional identification methods often involve invasive physical tags and can prove both disruptive for animals and time-intensive for researchers. In recent years, the integration of deep learning in research offered new methodological perspectives through automatization of complex tasks. Harnessing object detection and recognition technologies is increasingly used by researchers to achieve identification on video footage. This study represents a preliminary exploration into the development of a non-invasive tool for face detection and individual identification of Japanese macaques (Macaca fuscata) through deep learning. The ultimate goal of this research is, using identifications done on the dataset, to automatically generate a social network representation of the studied population. The current main results are promising: (i) the creation of a Japanese macaques' face detector (Faster-RCNN model), reaching a 82.2% accuracy and (ii) the creation of an individual recognizer for K{\=o}jima island macaques population (YOLOv8n model), reaching a 83% accuracy. We also created a K{\=o}jima population social network by traditional methods, based on co-occurrences on videos. Thus, we provide a benchmark against which the automatically generated network will be assessed for reliability. These preliminary results are a testament to the potential of this innovative approach to provide the scientific community with a tool for tracking individuals and social network studies in Japanese macaques.
Characterization of the Complexity of Computing the Capacity of Colored Noise Gaussian Channels
Authors: Holger Boche, Andrea Grigorescu, Rafael F. Schaefer, H. Vincent Poor
Abstract
This paper explores the computational complexity involved in determining the capacity of the band-limited additive colored Gaussian noise (ACGN) channel and its capacity-achieving power spectral density (p.s.d.). The study reveals that when the noise p.s.d. is a strictly positive computable continuous function, computing the capacity of the band-limited ACGN channel becomes a $#\mathrm{P}_1$-complete problem within the set of polynomial time computable noise p.s.d.s. Meaning that it is even more complex than problems that are $\mathrm{NP}_1$-complete. Additionally, it is shown that the capacity-achieving distribution is also $#\mathrm{P}_1$-complete. Furthermore, under the widely accepted assumption that $\mathrm{FP}_1 \neq #\mathrm{P}_1$, it has two significant implications for the ACGN channel. The first implication is the existence of a polynomial time computable noise p.s.d. for which the computation of its capacity cannot be performed in polynomial time, i.e., the number of computational steps on a Turing Machine grows faster than all polynomials. The second one is the existence of a polynomial time computable noise p.s.d. for which determining its capacity-achieving p.s.d. cannot be done within polynomial time.
Hierarchical Mask2Former: Panoptic Segmentation of Crops, Weeds and Leaves
Authors: Madeleine Darbyshire, Elizabeth Sklar, Simon Parsons
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Advancements in machine vision that enable detailed inferences to be made from images have the potential to transform many sectors including agriculture. Precision agriculture, where data analysis enables interventions to be precisely targeted, has many possible applications. Precision spraying, for example, can limit the application of herbicide only to weeds, or limit the application of fertiliser only to undernourished crops, instead of spraying the entire field. The approach promises to maximise yields, whilst minimising resource use and harms to the surrounding environment. To this end, we propose a hierarchical panoptic segmentation method to simultaneously identify indicators of plant growth and locate weeds within an image. We adapt Mask2Former, a state-of-the-art architecture for panoptic segmentation, to predict crop, weed and leaf masks. We achieve a PQ{\dag} of 75.99. Additionally, we explore approaches to make the architecture more compact and therefore more suitable for time and compute constrained applications. With our more compact architecture, inference is up to 60% faster and the reduction in PQ{\dag} is less than 1%.
FTFT: efficient and robust Fine-Tuning by transFerring Training dynamics
Authors: Yupei Du, Albert Gatt, Dong Nguyen
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Abstract
Despite the massive success of fine-tuning large Pre-trained Language Models (PLMs) on a wide range of Natural Language Processing (NLP) tasks, they remain susceptible to out-of-distribution (OOD) and adversarial inputs. Data map (DM) is a simple yet effective dual-model approach that enhances the robustness of fine-tuned PLMs, which involves fine-tuning a model on the original training set (i.e. reference model), selecting a specified fraction of important training examples according to the training dynamics of the reference model, and fine-tuning the same model on these selected examples (i.e. main model). However, it suffers from the drawback of requiring fine-tuning the same model twice, which is computationally expensive for large models. In this paper, we first show that 1) training dynamics are highly transferable across different model sizes and different pre-training methods, and that 2) main models fine-tuned using DM learn faster than when using conventional Empirical Risk Minimization (ERM). Building on these observations, we propose a novel fine-tuning approach based on the DM method: Fine-Tuning by transFerring Training dynamics (FTFT). Compared with DM, FTFT uses more efficient reference models and then fine-tunes more capable main models for fewer steps. Our experiments show that FTFT achieves better generalization robustness than ERM while spending less than half of the training cost.
EViT: An Eagle Vision Transformer with Bi-Fovea Self-Attention
Abstract
Because of the advancement of deep learning technology, vision transformer has demonstrated competitive performance in various computer vision tasks. Unfortunately, vision transformer still faces some challenges such as high computational complexity and absence of desirable inductive bias. To alleviate these problems, this study proposes a novel Bi-Fovea Self-Attention (BFSA) inspired by the physiological structure and characteristics of bi-fovea vision in eagle eyes. This BFSA can simulate the shallow fovea and deep fovea functions of eagle vision, enabling the network to extract feature representations of targets from coarse to fine, facilitating the interaction of multi-scale feature representations. Additionally, this study designs a Bionic Eagle Vision (BEV) block based on BFSA and CNN. It combines CNN and Vision Transformer, to enhance the network's local and global representation ability for targets. Furthermore, this study develops a unified and efficient general pyramid backbone network family, named Eagle Vision Transformers (EViTs) by stacking the BEV blocks. Experimental results on various computer vision tasks including image classification, object detection, instance segmentation and other transfer learning tasks show that the proposed EViTs perform significantly better than the baselines under similar model sizes, which exhibits faster speed on graphics processing unit compared to other models. Code will be released at https://github.com/nkusyl.
Mistral 7B
Authors: Albert Q. Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, Lélio Renard Lavaud, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Abstract
We introduce Mistral 7B v0.1, a 7-billion-parameter language model engineered for superior performance and efficiency. Mistral 7B outperforms Llama 2 13B across all evaluated benchmarks, and Llama 1 34B in reasoning, mathematics, and code generation. Our model leverages grouped-query attention (GQA) for faster inference, coupled with sliding window attention (SWA) to effectively handle sequences of arbitrary length with a reduced inference cost. We also provide a model fine-tuned to follow instructions, Mistral 7B -- Instruct, that surpasses the Llama 2 13B -- Chat model both on human and automated benchmarks. Our models are released under the Apache 2.0 license.
Keyword: mobile
Deep Learning based Tomato Disease Detection and Remedy Suggestions using Mobile Application
Abstract
We have developed a comprehensive computer system to assist farmers who practice traditional farming methods and have limited access to agricultural experts for addressing crop diseases. Our system utilizes artificial intelligence (AI) to identify and provide remedies for vegetable diseases. To ensure ease of use, we have created a mobile application that offers a user-friendly interface, allowing farmers to inquire about vegetable diseases and receive suitable solutions in their local language. The developed system can be utilized by any farmer with a basic understanding of a smartphone. Specifically, we have designed an AI-enabled mobile application for identifying and suggesting remedies for vegetable diseases, focusing on tomato diseases to benefit the local farming community in Nepal. Our system employs state-of-the-art object detection methodology, namely You Only Look Once (YOLO), to detect tomato diseases. The detected information is then relayed to the mobile application, which provides remedy suggestions guided by domain experts. In order to train our system effectively, we curated a dataset consisting of ten classes of tomato diseases. We utilized various data augmentation methods to address overfitting and trained a YOLOv5 object detector. The proposed method achieved a mean average precision of 0.76 and offers an efficient mobile interface for interacting with the AI system. While our system is currently in the development phase, we are actively working towards enhancing its robustness and real-time usability by accumulating more training samples.
Review of control algorithms for mobile robotics
Authors: Andres-David Suarez-Gomez, Andres A. Hernandez Ortega
Abstract
This article presents a comprehensive review of control algorithms used in mobile robotics, a field in constant evolution. Mobile robotics has seen significant advances in recent years, driven by the demand for applications in various sectors, such as industrial automation, space exploration, and medical care. The review focuses on control algorithms that address specific challenges in navigation, localization, mapping, and path planning in changing and unknown environments. Classical approaches, such as PID control and methods based on classical control theory, as well as modern techniques, including deep learning and model-based planning, are discussed in detail. In addition, practical applications and remaining challenges in implementing these algorithms in real-world mobile robots are highlighted. Ultimately, this review provides a comprehensive overview of the diversity and complexity of control algorithms in mobile robotics, helping researchers and practitioners to better understand the options available to address specific problems in this exciting area of study.
Layout Sequence Prediction From Noisy Mobile Modality
Authors: Haichao Zhang, Yi Xu, Hongsheng Lu, Takayuki Shimizu, Yun Fu
Abstract
Trajectory prediction plays a vital role in understanding pedestrian movement for applications such as autonomous driving and robotics. Current trajectory prediction models depend on long, complete, and accurately observed sequences from visual modalities. Nevertheless, real-world situations often involve obstructed cameras, missed objects, or objects out of sight due to environmental factors, leading to incomplete or noisy trajectories. To overcome these limitations, we propose LTrajDiff, a novel approach that treats objects obstructed or out of sight as equally important as those with fully visible trajectories. LTrajDiff utilizes sensor data from mobile phones to surmount out-of-sight constraints, albeit introducing new challenges such as modality fusion, noisy data, and the absence of spatial layout and object size information. We employ a denoising diffusion model to predict precise layout sequences from noisy mobile data using a coarse-to-fine diffusion strategy, incorporating the RMS, Siamese Masked Encoding Module, and MFM. Our model predicts layout sequences by implicitly inferring object size and projection status from a single reference timestamp or significantly obstructed sequences. Achieving SOTA results in randomly obstructed experiments and extremely short input experiments, our model illustrates the effectiveness of leveraging noisy mobile data. In summary, our approach offers a promising solution to the challenges faced by layout sequence and trajectory prediction models in real-world settings, paving the way for utilizing sensor data from mobile phones to accurately predict pedestrian bounding box trajectories. To the best of our knowledge, this is the first work that addresses severely obstructed and extremely short layout sequences by combining vision with noisy mobile modality, making it the pioneering work in the field of layout sequence trajectory prediction.
High-Fidelity 3D Head Avatars Reconstruction through Spatially-Varying Expression Conditioned Neural Radiance Field
Abstract
One crucial aspect of 3D head avatar reconstruction lies in the details of facial expressions. Although recent NeRF-based photo-realistic 3D head avatar methods achieve high-quality avatar rendering, they still encounter challenges retaining intricate facial expression details because they overlook the potential of specific expression variations at different spatial positions when conditioning the radiance field. Motivated by this observation, we introduce a novel Spatially-Varying Expression (SVE) conditioning. The SVE can be obtained by a simple MLP-based generation network, encompassing both spatial positional features and global expression information. Benefiting from rich and diverse information of the SVE at different positions, the proposed SVE-conditioned neural radiance field can deal with intricate facial expressions and achieve realistic rendering and geometry details of high-fidelity 3D head avatars. Additionally, to further elevate the geometric and rendering quality, we introduce a new coarse-to-fine training strategy, including a geometry initialization strategy at the coarse stage and an adaptive importance sampling strategy at the fine stage. Extensive experiments indicate that our method outperforms other state-of-the-art (SOTA) methods in rendering and geometry quality on mobile phone-collected and public datasets.
MEC-Intelligent Agent Support for Low-Latency Data Plane in Private NextG Core
Abstract
Private 5G networks will soon be ubiquitous across the future-generation smart wireless access infrastructures hosting a wide range of performance-critical applications. A high-performing User Plane Function (UPF) in the data plane is critical to achieving such stringent performance goals, as it governs fast packet processing and supports several key control-plane operations. Based on a private 5G prototype implementation and analysis, it is imperative to perform dynamic resource management and orchestration at the UPF. This paper leverages Mobile Edge Cloud-Intelligent Agent (MEC-IA), a logically centralized entity that proactively distributes resources at UPF for various service types, significantly reducing the tail latency experienced by the user requests while maximizing resource utilization. Extending the MEC-IA functionality to MEC layers further incurs data plane latency reduction. Based on our extensive simulations, under skewed uRLLC traffic arrival, the MEC-IA assisted bestfit UPF-MEC scheme reduces the worst-case latency of UE requests by up to 77.8% w.r.t. baseline. Additionally, the system can increase uRLLC connectivity gain by 2.40x while obtaining 40% CapEx savings.
Redundant and Loosely Coupled LiDAR-Wi-Fi Integration for Robust Global Localization in Autonomous Mobile Robotics
Authors: Nikolaos Stathoulopoulos, Emanuele Pagliari, Luca Davoli, George Nikolakopoulos
Abstract
This paper presents a framework addressing the challenge of global localization in autonomous mobile robotics by integrating LiDAR-based descriptors and Wi-Fi fingerprinting in a pre-mapped environment. This is motivated by the increasing demand for reliable localization in complex scenarios, such as urban areas or underground mines, requiring robust systems able to overcome limitations faced by traditional Global Navigation Satellite System (GNSS)-based localization methods. By leveraging the complementary strengths of LiDAR and Wi-Fi sensors used to generate predictions and evaluate the confidence of each prediction as an indicator of potential degradation, we propose a redundancy-based approach that enhances the system's overall robustness and accuracy. The proposed framework allows independent operation of the LiDAR and Wi-Fi sensors, ensuring system redundancy. By combining the predictions while considering their confidence levels, we achieve enhanced and consistent performance in localization tasks.
A Supervised Embedding and Clustering Anomaly Detection method for classification of Mobile Network Faults
Abstract
The paper introduces Supervised Embedding and Clustering Anomaly Detection (SEMC-AD), a method designed to efficiently identify faulty alarm logs in a mobile network and alleviate the challenges of manual monitoring caused by the growing volume of alarm logs. SEMC-AD employs a supervised embedding approach based on deep neural networks, utilizing historical alarm logs and their labels to extract numerical representations for each log, effectively addressing the issue of imbalanced classification due to a small proportion of anomalies in the dataset without employing one-hot encoding. The robustness of the embedding is evaluated by plotting the two most significant principle components of the embedded alarm logs, revealing that anomalies form distinct clusters with similar embeddings. Multivariate normal Gaussian clustering is then applied to these components, identifying clusters with a high ratio of anomalies to normal alarms (above 90%) and labeling them as the anomaly group. To classify new alarm logs, we check if their embedded vectors' two most significant principle components fall within the anomaly-labeled clusters. If so, the log is classified as an anomaly. Performance evaluation demonstrates that SEMC-AD outperforms conventional random forest and gradient boosting methods without embedding. SEMC-AD achieves 99% anomaly detection, whereas random forest and XGBoost only detect 86% and 81% of anomalies, respectively. While supervised classification methods may excel in labeled datasets, the results demonstrate that SEMC-AD is more efficient in classifying anomalies in datasets with numerous categorical features, significantly enhancing anomaly detection, reducing operator burden, and improving network maintenance.
Keyword: pruning
Compressing Context to Enhance Inference Efficiency of Large Language Models
Authors: Yucheng Li, Bo Dong, Chenghua Lin, Frank Guerin
Abstract
Large language models (LLMs) achieved remarkable performance across various tasks. However, they face challenges in managing long documents and extended conversations, due to significantly increased computational requirements, both in memory and inference time, and potential context truncation when the input exceeds the LLM's fixed context length. This paper proposes a method called Selective Context that enhances the inference efficiency of LLMs by identifying and pruning redundancy in the input context to make the input more compact. We test our approach using common data sources requiring long context processing: arXiv papers, news articles, and long conversations, on tasks of summarisation, question answering, and response generation. Experimental results show that Selective Context significantly reduces memory cost and decreases generation latency while maintaining comparable performance compared to that achieved when full context is used. Specifically, we achieve a 50\% reduction in context cost, resulting in a 36\% reduction in inference memory usage and a 32\% reduction in inference time, while observing only a minor drop of .023 in BERTscore and .038 in faithfulness on four downstream applications, indicating that our method strikes a good balance between efficiency and performance.
Abstract
The study of sparsity in Convolutional Neural Networks (CNNs) has become widespread to compress and accelerate models in environments with limited resources. By constraining N consecutive weights along the output channel to be group-wise non-zero, the recent network with 1$\times$N sparsity has received tremendous popularity for its three outstanding advantages: 1) A large amount of storage space saving by a \emph{Block Sparse Row} matrix. 2) Excellent performance at a high sparsity. 3) Significant speedups on CPUs with Advanced Vector Extensions. Recent work requires selecting and fine-tuning 1$\times$N sparse weights based on dense pre-trained weights, leading to the problems such as expensive training cost and memory access, sub-optimal model quality, as well as unbalanced workload across threads (different sparsity across output channels). To overcome them, this paper proposes a novel \emph{\textbf{S}oft \textbf{U}niform \textbf{B}lock \textbf{P}runing} (SUBP) approach to train a uniform 1$\times$N sparse structured network from scratch. Specifically, our approach tends to repeatedly allow pruned blocks to regrow to the network based on block angular redundancy and importance sampling in a uniform manner throughout the training process. It not only makes the model less dependent on pre-training, reduces the model redundancy and the risk of pruning the important blocks permanently but also achieves balanced workload. Empirically, on ImageNet, comprehensive experiments across various CNN architectures show that our SUBP consistently outperforms existing 1$\times$N and structured sparsity methods based on pre-trained models or training from scratch. Source codes and models are available at \url{https://github.com/JingyangXiang/SUBP}.
Filter Pruning For CNN With Enhanced Linear Representation Redundancy
Abstract
Structured network pruning excels non-structured methods because they can take advantage of the thriving developed parallel computing techniques. In this paper, we propose a new structured pruning method. Firstly, to create more structured redundancy, we present a data-driven loss function term calculated from the correlation coefficient matrix of different feature maps in the same layer, named CCM-loss. This loss term can encourage the neural network to learn stronger linear representation relations between feature maps during the training from the scratch so that more homogenous parts can be removed later in pruning. CCM-loss provides us with another universal transcendental mathematical tool besides L*-norm regularization, which concentrates on generating zeros, to generate more redundancy but for the different genres. Furthermore, we design a matching channel selection strategy based on principal components analysis to exploit the maximum potential ability of CCM-loss. In our new strategy, we mainly focus on the consistency and integrality of the information flow in the network. Instead of empirically hard-code the retain ratio for each layer, our channel selection strategy can dynamically adjust each layer's retain ratio according to the specific circumstance of a per-trained model to push the prune ratio to the limit. Notably, on the Cifar-10 dataset, our method brings 93.64% accuracy for pruned VGG-16 with only 1.40M parameters and 49.60M FLOPs, the pruned ratios for parameters and FLOPs are 90.6% and 84.2%, respectively. For ResNet-50 trained on the ImageNet dataset, our approach achieves 42.8% and 47.3% storage and computation reductions, respectively, with an accuracy of 76.23%. Our code is available at https://github.com/Bojue-Wang/CCM-LRR.
Skeleton Ground Truth Extraction: Methodology, Annotation Tool and Benchmarks
Authors: Cong Yang, Bipin Indurkhya, John See, Bo Gao, Yan Ke, Zeyd Boukhers, Zhenyu Yang, Marcin Grzegorzek
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Abstract
Skeleton Ground Truth (GT) is critical to the success of supervised skeleton extraction methods, especially with the popularity of deep learning techniques. Furthermore, we see skeleton GTs used not only for training skeleton detectors with Convolutional Neural Networks (CNN) but also for evaluating skeleton-related pruning and matching algorithms. However, most existing shape and image datasets suffer from the lack of skeleton GT and inconsistency of GT standards. As a result, it is difficult to evaluate and reproduce CNN-based skeleton detectors and algorithms on a fair basis. In this paper, we present a heuristic strategy for object skeleton GT extraction in binary shapes and natural images. Our strategy is built on an extended theory of diagnosticity hypothesis, which enables encoding human-in-the-loop GT extraction based on clues from the target's context, simplicity, and completeness. Using this strategy, we developed a tool, SkeView, to generate skeleton GT of 17 existing shape and image datasets. The GTs are then structurally evaluated with representative methods to build viable baselines for fair comparisons. Experiments demonstrate that GTs generated by our strategy yield promising quality with respect to standard consistency, and also provide a balance between simplicity and completeness.
Abstract
Machine learning models need to be continually updated or corrected to ensure that the prediction accuracy remains consistently high. In this study, we consider scenarios where developers should be careful to change the prediction results by the model correction, such as when the model is part of a complex system or software. In such scenarios, the developers want to control the specification of the corrections. To achieve this, the developers need to understand which subpopulations of the inputs get inaccurate predictions by the model. Therefore, we propose correction rule mining to acquire a comprehensive list of rules that describe inaccurate subpopulations and how to correct them. We also develop an efficient correction rule mining algorithm that is a combination of frequent itemset mining and a unique pruning technique for correction rules. We observed that the proposed algorithm found various rules which help to collect data insufficiently learned, directly correct model outputs, and analyze concept drift.
Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning
Abstract
The popularity of LLaMA (Touvron et al., 2023a;b) and other recently emerged moderate-sized large language models (LLMs) highlights the potential of building smaller yet powerful LLMs. Regardless, the cost of training such models from scratch on trillions of tokens remains high. In this work, we study structured pruning as an effective means to develop smaller LLMs from pre-trained, larger models. Our approach employs two key techniques: (1) targeted structured pruning, which prunes a larger model to a specified target shape by removing layers, heads, and intermediate and hidden dimensions in an end-to-end manner, and (2) dynamic batch loading, which dynamically updates the composition of sampled data in each training batch based on varying losses across different domains. We demonstrate the efficacy of our approach by presenting the Sheared-LLaMA series, pruning the LLaMA2-7B model down to 1.3B and 2.7B parameters. Sheared-LLaMA models outperform state-of-the-art open-source models of equivalent sizes, such as Pythia, INCITE, and OpenLLaMA models, on a wide range of downstream and instruction tuning evaluations, while requiring only 3% of compute compared to training such models from scratch. This work provides compelling evidence that leveraging existing LLMs with structured pruning is a far more cost-effective approach for building smaller LLMs.
Keyword: diffusion
DF-3DFace: One-to-Many Speech Synchronized 3D Face Animation with Diffusion
Authors: Se Jin Park, Joanna Hong, Minsu Kim, Yong Man Ro
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Multimedia (cs.MM); Image and Video Processing (eess.IV)
Abstract
Speech-driven 3D facial animation has gained significant attention for its ability to create realistic and expressive facial animations in 3D space based on speech. Learning-based methods have shown promising progress in achieving accurate facial motion synchronized with speech. However, one-to-many nature of speech-to-3D facial synthesis has not been fully explored: while the lip accurately synchronizes with the speech content, other facial attributes beyond speech-related motions are variable with respect to the speech. To account for the potential variance in the facial attributes within a single speech, we propose DF-3DFace, a diffusion-driven speech-to-3D face mesh synthesis. DF-3DFace captures the complex one-to-many relationships between speech and 3D face based on diffusion. It concurrently achieves aligned lip motion by exploiting audio-mesh synchronization and masked conditioning. Furthermore, the proposed method jointly models identity and pose in addition to facial motions so that it can generate 3D face animation without requiring a reference identity mesh and produce natural head poses. We contribute a new large-scale 3D facial mesh dataset, 3D-HDTF to enable the synthesis of variations in identities, poses, and facial motions of 3D face mesh. Extensive experiments demonstrate that our method successfully generates highly variable facial shapes and motions from speech and simultaneously achieves more realistic facial animation than the state-of-the-art methods.
Stabilized finite elements for the solution of the Reynolds equation considering cavitation
Authors: Hauke Gravenkamp, Simon Pfeil, Ramon Codina
Abstract
The Reynolds equation, combined with the Elrod algorithm for including the effect of cavitation, resembles a nonlinear convection-diffusion-reaction (CDR) equation. Its solution by finite elements is prone to oscillations in convection-dominated regions, which are present whenever cavitation occurs. We propose a stabilized finite-element method that is based on the variational multiscale method and exploits the concept of orthogonal subgrid scales. We demonstrate that this approach only requires one additional term in the weak form to obtain a stable method that converges optimally when performing mesh refinement.
Layout Sequence Prediction From Noisy Mobile Modality
Authors: Haichao Zhang, Yi Xu, Hongsheng Lu, Takayuki Shimizu, Yun Fu
Abstract
Trajectory prediction plays a vital role in understanding pedestrian movement for applications such as autonomous driving and robotics. Current trajectory prediction models depend on long, complete, and accurately observed sequences from visual modalities. Nevertheless, real-world situations often involve obstructed cameras, missed objects, or objects out of sight due to environmental factors, leading to incomplete or noisy trajectories. To overcome these limitations, we propose LTrajDiff, a novel approach that treats objects obstructed or out of sight as equally important as those with fully visible trajectories. LTrajDiff utilizes sensor data from mobile phones to surmount out-of-sight constraints, albeit introducing new challenges such as modality fusion, noisy data, and the absence of spatial layout and object size information. We employ a denoising diffusion model to predict precise layout sequences from noisy mobile data using a coarse-to-fine diffusion strategy, incorporating the RMS, Siamese Masked Encoding Module, and MFM. Our model predicts layout sequences by implicitly inferring object size and projection status from a single reference timestamp or significantly obstructed sequences. Achieving SOTA results in randomly obstructed experiments and extremely short input experiments, our model illustrates the effectiveness of leveraging noisy mobile data. In summary, our approach offers a promising solution to the challenges faced by layout sequence and trajectory prediction models in real-world settings, paving the way for utilizing sensor data from mobile phones to accurately predict pedestrian bounding box trajectories. To the best of our knowledge, this is the first work that addresses severely obstructed and extremely short layout sequences by combining vision with noisy mobile modality, making it the pioneering work in the field of layout sequence trajectory prediction.
Latent Diffusion Model for DNA Sequence Generation
Authors: Zehui Li, Yuhao Ni, Tim August B. Huygelen, Akashaditya Das, Guoxuan Xia, Guy-Bart Stan, Yiren Zhao
Abstract
The harnessing of machine learning, especially deep generative models, has opened up promising avenues in the field of synthetic DNA sequence generation. Whilst Generative Adversarial Networks (GANs) have gained traction for this application, they often face issues such as limited sample diversity and mode collapse. On the other hand, Diffusion Models are a promising new class of generative models that are not burdened with these problems, enabling them to reach the state-of-the-art in domains such as image generation. In light of this, we propose a novel latent diffusion model, DiscDiff, tailored for discrete DNA sequence generation. By simply embedding discrete DNA sequences into a continuous latent space using an autoencoder, we are able to leverage the powerful generative abilities of continuous diffusion models for the generation of discrete data. Additionally, we introduce Fr\'echet Reconstruction Distance (FReD) as a new metric to measure the sample quality of DNA sequence generations. Our DiscDiff model demonstrates an ability to generate synthetic DNA sequences that align closely with real DNA in terms of Motif Distribution, Latent Embedding Distribution (FReD), and Chromatin Profiles. Additionally, we contribute a comprehensive cross-species dataset of 150K unique promoter-gene sequences from 15 species, enriching resources for future generative modelling in genomics. We will make our code public upon publication.
Memory-Consistent Neural Networks for Imitation Learning
Authors: Kaustubh Sridhar, Souradeep Dutta, Dinesh Jayaraman, James Weimer, Insup Lee
Abstract
Imitation learning considerably simplifies policy synthesis compared to alternative approaches by exploiting access to expert demonstrations. For such imitation policies, errors away from the training samples are particularly critical. Even rare slip-ups in the policy action outputs can compound quickly over time, since they lead to unfamiliar future states where the policy is still more likely to err, eventually causing task failures. We revisit simple supervised behavior cloning'' for conveniently training the policy from nothing more than pre-recorded demonstrations, but carefully design the model class to counter the compounding error phenomenon. Ourmemory-consistent neural network'' (MCNN) outputs are hard-constrained to stay within clearly specified permissible regions anchored to prototypical ``memory'' training samples. We provide a guaranteed upper bound for the sub-optimality gap induced by MCNN policies. Using MCNNs on 9 imitation learning tasks, with MLP, Transformer, and Diffusion backbones, spanning dexterous robotic manipulation and driving, proprioceptive inputs and visual inputs, and varying sizes and types of demonstration data, we find large and consistent gains in performance, validating that MCNNs are better-suited than vanilla deep neural networks for imitation learning applications. Website: https://sites.google.com/view/mcnn-imitation
DockGame: Cooperative Games for Multimeric Rigid Protein Docking
Authors: Vignesh Ram Somnath, Pier Giuseppe Sessa, Maria Rodriguez Martinez, Andreas Krause
Abstract
Protein interactions and assembly formation are fundamental to most biological processes. Predicting the assembly structure from constituent proteins -- referred to as the protein docking task -- is thus a crucial step in protein design applications. Most traditional and deep learning methods for docking have focused mainly on binary docking, following either a search-based, regression-based, or generative modeling paradigm. In this paper, we focus on the less-studied multimeric (i.e., two or more proteins) docking problem. We introduce DockGame, a novel game-theoretic framework for docking -- we view protein docking as a cooperative game between proteins, where the final assembly structure(s) constitute stable equilibria w.r.t. the underlying game potential. Since we do not have access to the true potential, we consider two approaches - i) learning a surrogate game potential guided by physics-based energy functions and computing equilibria by simultaneous gradient updates, and ii) sampling from the Gibbs distribution of the true potential by learning a diffusion generative model over the action spaces (rotations and translations) of all proteins. Empirically, on the Docking Benchmark 5.5 (DB5.5) dataset, DockGame has much faster runtimes than traditional docking methods, can generate multiple plausible assembly structures, and achieves comparable performance to existing binary docking baselines, despite solving the harder task of coordinating multiple protein chains.
Improving Compositional Text-to-image Generation with Large Vision-Language Models
Abstract
Recent advancements in text-to-image models, particularly diffusion models, have shown significant promise. However, compositional text-to-image models frequently encounter difficulties in generating high-quality images that accurately align with input texts describing multiple objects, variable attributes, and intricate spatial relationships. To address this limitation, we employ large vision-language models (LVLMs) for multi-dimensional assessment of the alignment between generated images and their corresponding input texts. Utilizing this assessment, we fine-tune the diffusion model to enhance its alignment capabilities. During the inference phase, an initial image is produced using the fine-tuned diffusion model. The LVLM is then employed to pinpoint areas of misalignment in the initial image, which are subsequently corrected using the image editing algorithm until no further misalignments are detected by the LVLM. The resultant image is consequently more closely aligned with the input text. Our experimental results validate that the proposed methodology significantly improves text-image alignment in compositional image generation, particularly with respect to object number, attribute binding, spatial relationships, and aesthetic quality.
Advancing Pose-Guided Image Synthesis with Progressive Conditional Diffusion Models
Authors: Fei Shen, Hu Ye, Jun Zhang, Cong Wang, Xiao Han, Wei Yang
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent work has showcased the significant potential of diffusion models in pose-guided person image synthesis. However, owing to the inconsistency in pose between the source and target images, synthesizing an image with a distinct pose, relying exclusively on the source image and target pose information, remains a formidable challenge. This paper presents Progressive Conditional Diffusion Models (PCDMs) that incrementally bridge the gap between person images under the target and source poses through three stages. Specifically, in the first stage, we design a simple prior conditional diffusion model that predicts the global features of the target image by mining the global alignment relationship between pose coordinates and image appearance. Then, the second stage establishes a dense correspondence between the source and target images using the global features from the previous stage, and an inpainting conditional diffusion model is proposed to further align and enhance the contextual features, generating a coarse-grained person image. In the third stage, we propose a refining conditional diffusion model to utilize the coarsely generated image from the previous stage as a condition, achieving texture restoration and enhancing fine-detail consistency. The three-stage PCDMs work progressively to generate the final high-quality and high-fidelity synthesized image. Both qualitative and quantitative results demonstrate the consistency and photorealism of our proposed PCDMs under challenging scenarios.The code and model will be available at https://github.com/muzishen/PCDMs.
Boosting Continuous Control with Consistency Policy
Abstract
Due to its training stability and strong expression, the diffusion model has attracted considerable attention in offline reinforcement learning. However, several challenges have also come with it: 1) The demand for a large number of diffusion steps makes the diffusion-model-based methods time inefficient and limits their applications in real-time control; 2) How to achieve policy improvement with accurate guidance for diffusion model-based policy is still an open problem. Inspired by the consistency model, we propose a novel time-efficiency method named Consistency Policy with Q-Learning (CPQL), which derives action from noise by a single step. By establishing a mapping from the reverse diffusion trajectories to the desired policy, we simultaneously address the issues of time efficiency and inaccurate guidance when updating diffusion model-based policy with the learned Q-function. We demonstrate that CPQL can achieve policy improvement with accurate guidance for offline reinforcement learning, and can be seamlessly extended for online RL tasks. Experimental results indicate that CPQL achieves new state-of-the-art performance on 11 offline and 21 online tasks, significantly improving inference speed by nearly 45 times compared to Diffusion-QL. We will release our code later.
JointNet: Extending Text-to-Image Diffusion for Dense Distribution Modeling
Authors: Jingyang Zhang, Shiwei Li, Yuanxun Lu, Tian Fang, David McKinnon, Yanghai Tsin, Long Quan, Yao Yao
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
We introduce JointNet, a novel neural network architecture for modeling the joint distribution of images and an additional dense modality (e.g., depth maps). JointNet is extended from a pre-trained text-to-image diffusion model, where a copy of the original network is created for the new dense modality branch and is densely connected with the RGB branch. The RGB branch is locked during network fine-tuning, which enables efficient learning of the new modality distribution while maintaining the strong generalization ability of the large-scale pre-trained diffusion model. We demonstrate the effectiveness of JointNet by using RGBD diffusion as an example and through extensive experiments, showcasing its applicability in a variety of applications, including joint RGBD generation, dense depth prediction, depth-conditioned image generation, and coherent tile-based 3D panorama generation.
Leveraging Diffusion-Based Image Variations for Robust Training on Poisoned Data
Authors: Lukas Struppek, Martin B. Hentschel, Clifton Poth, Dominik Hintersdorf, Kristian Kersting
Subjects: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Abstract
Backdoor attacks pose a serious security threat for training neural networks as they surreptitiously introduce hidden functionalities into a model. Such backdoors remain silent during inference on clean inputs, evading detection due to inconspicuous behavior. However, once a specific trigger pattern appears in the input data, the backdoor activates, causing the model to execute its concealed function. Detecting such poisoned samples within vast datasets is virtually impossible through manual inspection. To address this challenge, we propose a novel approach that enables model training on potentially poisoned datasets by utilizing the power of recent diffusion models. Specifically, we create synthetic variations of all training samples, leveraging the inherent resilience of diffusion models to potential trigger patterns in the data. By combining this generative approach with knowledge distillation, we produce student models that maintain their general performance on the task while exhibiting robust resistance to backdoor triggers.
Learning Stackable and Skippable LEGO Bricks for Efficient, Reconfigurable, and Variable-Resolution Diffusion Modeling
Abstract
Diffusion models excel at generating photo-realistic images but come with significant computational costs in both training and sampling. While various techniques address these computational challenges, a less-explored issue is designing an efficient and adaptable network backbone for iterative refinement. Current options like U-Net and Vision Transformer often rely on resource-intensive deep networks and lack the flexibility needed for generating images at variable resolutions or with a smaller network than used in training. This study introduces LEGO bricks, which seamlessly integrate Local-feature Enrichment and Global-content Orchestration. These bricks can be stacked to create a test-time reconfigurable diffusion backbone, allowing selective skipping of bricks to reduce sampling costs and generate higher-resolution images than the training data. LEGO bricks enrich local regions with an MLP and transform them using a Transformer block while maintaining a consistent full-resolution image across all bricks. Experimental results demonstrate that LEGO bricks enhance training efficiency, expedite convergence, and facilitate variable-resolution image generation while maintaining strong generative performance. Moreover, LEGO significantly reduces sampling time compared to other methods, establishing it as a valuable enhancement for diffusion models.
Advective Diffusion Transformers for Topological Generalization in Graph Learning
Authors: Qitian Wu, Chenxiao Yang, Kaipeng Zeng, Fan Nie, Michael Bronstein, Junchi Yan
Abstract
Graph diffusion equations are intimately related to graph neural networks (GNNs) and have recently attracted attention as a principled framework for analyzing GNN dynamics, formalizing their expressive power, and justifying architectural choices. One key open questions in graph learning is the generalization capabilities of GNNs. A major limitation of current approaches hinges on the assumption that the graph topologies in the training and test sets come from the same distribution. In this paper, we make steps towards understanding the generalization of GNNs by exploring how graph diffusion equations extrapolate and generalize in the presence of varying graph topologies. We first show deficiencies in the generalization capability of existing models built upon local diffusion on graphs, stemming from the exponential sensitivity to topology variation. Our subsequent analysis reveals the promise of non-local diffusion, which advocates for feature propagation over fully-connected latent graphs, under the assumption of a specific data-generating condition. In addition to these findings, we propose a novel graph encoder backbone, Advective Diffusion Transformer (ADiT), inspired by advective graph diffusion equations that have a closed-form solution backed up with theoretical guarantees of desired generalization under topological distribution shifts. The new model, functioning as a versatile graph Transformer, demonstrates superior performance across a wide range of graph learning tasks.
AnoDODE: Anomaly Detection with Diffusion ODE
Authors: Xianyao Hu, Congming Jin
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Anomaly detection is the process of identifying atypical data samples that significantly deviate from the majority of the dataset. In the realm of clinical screening and diagnosis, detecting abnormalities in medical images holds great importance. Typically, clinical practice provides access to a vast collection of normal images, while abnormal images are relatively scarce. We hypothesize that abnormal images and their associated features tend to manifest in low-density regions of the data distribution. Following this assumption, we turn to diffusion ODEs for unsupervised anomaly detection, given their tractability and superior performance in density estimation tasks. More precisely, we propose a new anomaly detection method based on diffusion ODEs by estimating the density of features extracted from multi-scale medical images. Our anomaly scoring mechanism depends on computing the negative log-likelihood of features extracted from medical images at different scales, quantified in bits per dimension. Furthermore, we propose a reconstruction-based anomaly localization suitable for our method. Our proposed method not only identifie anomalies but also provides interpretability at both the image and pixel levels. Through experiments on the BraTS2021 medical dataset, our proposed method outperforms existing methods. These results confirm the effectiveness and robustness of our method.
Tertiary Lymphoid Structures Generation through Graph-based Diffusion
Authors: Manuel Madeira, Dorina Thanou, Pascal Frossard
Abstract
Graph-based representation approaches have been proven to be successful in the analysis of biomedical data, due to their capability of capturing intricate dependencies between biological entities, such as the spatial organization of different cell types in a tumor tissue. However, to further enhance our understanding of the underlying governing biological mechanisms, it is important to accurately capture the actual distributions of such complex data. Graph-based deep generative models are specifically tailored to accomplish that. In this work, we leverage state-of-the-art graph-based diffusion models to generate biologically meaningful cell-graphs. In particular, we show that the adopted graph diffusion model is able to accurately learn the distribution of cells in terms of their tertiary lymphoid structures (TLS) content, a well-established biomarker for evaluating the cancer progression in oncology research. Additionally, we further illustrate the utility of the learned generative models for data augmentation in a TLS classification task. To the best of our knowledge, this is the first work that leverages the power of graph diffusion models in generating meaningful biological cell structures.
Latent Diffusion Counterfactual Explanations
Authors: Karim Farid, Simon Schrodi, Max Argus, Thomas Brox
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Counterfactual explanations have emerged as a promising method for elucidating the behavior of opaque black-box models. Recently, several works leveraged pixel-space diffusion models for counterfactual generation. To handle noisy, adversarial gradients during counterfactual generation -- causing unrealistic artifacts or mere adversarial perturbations -- they required either auxiliary adversarially robust models or computationally intensive guidance schemes. However, such requirements limit their applicability, e.g., in scenarios with restricted access to the model's training data. To address these limitations, we introduce Latent Diffusion Counterfactual Explanations (LDCE). LDCE harnesses the capabilities of recent class- or text-conditional foundation latent diffusion models to expedite counterfactual generation and focus on the important, semantic parts of the data. Furthermore, we propose a novel consensus guidance mechanism to filter out noisy, adversarial gradients that are misaligned with the diffusion model's implicit classifier. We demonstrate the versatility of LDCE across a wide spectrum of models trained on diverse datasets with different learning paradigms. Finally, we showcase how LDCE can provide insights into model errors, enhancing our understanding of black-box model behavior.
HiFi-123: Towards High-fidelity One Image to 3D Content Generation
Authors: Wangbo Yu, Li Yuan, Yan-Pei Cao, Xiangjun Gao, Xiaoyu Li, Long Quan, Ying Shan, Yonghong Tian
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent advances in text-to-image diffusion models have enabled 3D generation from a single image. However, current image-to-3D methods often produce suboptimal results for novel views, with blurred textures and deviations from the reference image, limiting their practical applications. In this paper, we introduce HiFi-123, a method designed for high-fidelity and multi-view consistent 3D generation. Our contributions are twofold: First, we propose a reference-guided novel view enhancement technique that substantially reduces the quality gap between synthesized and reference views. Second, capitalizing on the novel view enhancement, we present a novel reference-guided state distillation loss. When incorporated into the optimization-based image-to-3D pipeline, our method significantly improves 3D generation quality, achieving state-of-the-art performance. Comprehensive evaluations demonstrate the effectiveness of our approach over existing methods, both qualitatively and quantitatively.
What Does Stable Diffusion Know about the 3D Scene?
Authors: Guanqi Zhan, Chuanxia Zheng, Weidi Xie, Andrew Zisserman
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent advances in generative models like Stable Diffusion enable the generation of highly photo-realistic images. Our objective in this paper is to probe the diffusion network to determine to what extent it 'understands' different properties of the 3D scene depicted in an image. To this end, we make the following contributions: (i) We introduce a protocol to evaluate whether a network models a number of physical 'properties' of the 3D scene by probing for explicit features that represent these properties. The probes are applied on datasets of real images with annotations for the property. (ii) We apply this protocol to properties covering scene geometry, scene material, support relations, lighting, and view dependent measures. (iii) We find that Stable Diffusion is good at a number of properties including scene geometry, support relations, shadows and depth, but less performant for occlusion. (iv) We also apply the probes to other models trained at large-scale, including DINO and CLIP, and find their performance inferior to that of Stable Diffusion.
Keyword: adaptive
Enhancing Document-level Event Argument Extraction with Contextual Clues and Role Relevance
Authors: Wanlong Liu, Shaohuan Cheng, Dingyi Zeng, Hong Qu
Abstract
Document-level event argument extraction poses new challenges of long input and cross-sentence inference compared to its sentence-level counterpart. However, most prior works focus on capturing the relations between candidate arguments and the event trigger in each event, ignoring two crucial points: a) non-argument contextual clue information; b) the relevance among argument roles. In this paper, we propose a SCPRG (Span-trigger-based Contextual Pooling and latent Role Guidance) model, which contains two novel and effective modules for the above problem. The Span-Trigger-based Contextual Pooling(STCP) adaptively selects and aggregates the information of non-argument clue words based on the context attention weights of specific argument-trigger pairs from pre-trained model. The Role-based Latent Information Guidance (RLIG) module constructs latent role representations, makes them interact through role-interactive encoding to capture semantic relevance, and merges them into candidate arguments. Both STCP and RLIG introduce no more than 1% new parameters compared with the base model and can be easily applied to other event extraction models, which are compact and transplantable. Experiments on two public datasets show that our SCPRG outperforms previous state-of-the-art methods, with 1.13 F1 and 2.64 F1 improvements on RAMS and WikiEvents respectively. Further analyses illustrate the interpretability of our model.
A Natural Indirect Adaptive Controller for a Satellite-Mounted Manipulator
Authors: Jacopo Giordano, Angelo Cenedese, Andrea Serrani
Abstract
The work considers the design of an indirect adaptive controller for a satellite equipped with a robotic arm manipulating an object. Uncertainty on the manipulated object can considerably impact the overall behavior of the system. In addition, the dynamics of the actuators of the base satellite are non-linear and can be affected by malfunctioning. Neglecting these two phenomena may lead to excessive control effort or degrade performance. An indirect adaptive control approach is pursued, which allows consideration of relevant features of the actuators dynamics, such as loss of effectiveness. Furthermore, an adaptive law that preserves the physical consistency of the inertial parameters of the various rigid bodies comprising the system is employed. The performance and robustness of the controller are first analyzed and then validated in simulation.
CAT-RRT: Motion Planning that Admits Contact One Link at a Time
Abstract
Current motion planning approaches rely on binary collision checking to evaluate the validity of a state and thereby dictate where the robot is allowed to move. This approach leaves little room for robots to engage in contact with an object, as is often necessary when operating in densely cluttered spaces. In this work, we propose an alternative method that considers contact states as high-cost states that the robot should avoid but can traverse if necessary to complete a task. More specifically, we introduce Contact Admissible Transition-based Rapidly exploring Random Trees (CAT-RRT), a planner that uses a novel per-link cost heuristic to find a path by traversing high-cost obstacle regions. Through extensive testing, we find that state-of-the-art optimization planners tend to over-explore low-cost states, which leads to slow and inefficient convergence to contact regions. Conversely, CAT-RRT searches both low and high-cost regions simultaneously with an adaptive thresholding mechanism carried out at each robot link. This leads to paths with a balance between efficiency, path length, and contact cost.
Efficient Adaptation of Large Vision Transformer via Adapter Re-Composing
Authors: Wei Dong, Dawei Yan, Zhijun Lin, Peng Wang
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Abstract
The advent of high-capacity pre-trained models has revolutionized problem-solving in computer vision, shifting the focus from training task-specific models to adapting pre-trained models. Consequently, effectively adapting large pre-trained models to downstream tasks in an efficient manner has become a prominent research area. Existing solutions primarily concentrate on designing lightweight adapters and their interaction with pre-trained models, with the goal of minimizing the number of parameters requiring updates. In this study, we propose a novel Adapter Re-Composing (ARC) strategy that addresses efficient pre-trained model adaptation from a fresh perspective. Our approach considers the reusability of adaptation parameters and introduces a parameter-sharing scheme. Specifically, we leverage symmetric down-/up-projections to construct bottleneck operations, which are shared across layers. By learning low-dimensional re-scaling coefficients, we can effectively re-compose layer-adaptive adapters. This parameter-sharing strategy in adapter design allows us to significantly reduce the number of new parameters while maintaining satisfactory performance, thereby offering a promising approach to compress the adaptation cost. We conduct experiments on 24 downstream image classification tasks using various Vision Transformer variants to evaluate our method. The results demonstrate that our approach achieves compelling transfer learning performance with a reduced parameter count. Our code is available at \href{https://github.com/DavidYanAnDe/ARC}{https://github.com/DavidYanAnDe/ARC}.
A Unified View on Solving Objective Mismatch in Model-Based Reinforcement Learning
Authors: Ran Wei, Nathan Lambert, Anthony McDonald, Alfredo Garcia, Roberto Calandra
Abstract
Model-based Reinforcement Learning (MBRL) aims to make agents more sample-efficient, adaptive, and explainable by learning an explicit model of the environment. While the capabilities of MBRL agents have significantly improved in recent years, how to best learn the model is still an unresolved question. The majority of MBRL algorithms aim at training the model to make accurate predictions about the environment and subsequently using the model to determine the most rewarding actions. However, recent research has shown that model predictive accuracy is often not correlated with action quality, tracing the root cause to the \emph{objective mismatch} between accurate dynamics model learning and policy optimization of rewards. A number of interrelated solution categories to the objective mismatch problem have emerged as MBRL continues to mature as a research area. In this work, we provide an in-depth survey of these solution categories and propose a taxonomy to foster future research.
High-Fidelity 3D Head Avatars Reconstruction through Spatially-Varying Expression Conditioned Neural Radiance Field
Abstract
One crucial aspect of 3D head avatar reconstruction lies in the details of facial expressions. Although recent NeRF-based photo-realistic 3D head avatar methods achieve high-quality avatar rendering, they still encounter challenges retaining intricate facial expression details because they overlook the potential of specific expression variations at different spatial positions when conditioning the radiance field. Motivated by this observation, we introduce a novel Spatially-Varying Expression (SVE) conditioning. The SVE can be obtained by a simple MLP-based generation network, encompassing both spatial positional features and global expression information. Benefiting from rich and diverse information of the SVE at different positions, the proposed SVE-conditioned neural radiance field can deal with intricate facial expressions and achieve realistic rendering and geometry details of high-fidelity 3D head avatars. Additionally, to further elevate the geometric and rendering quality, we introduce a new coarse-to-fine training strategy, including a geometry initialization strategy at the coarse stage and an adaptive importance sampling strategy at the fine stage. Extensive experiments indicate that our method outperforms other state-of-the-art (SOTA) methods in rendering and geometry quality on mobile phone-collected and public datasets.
Conformal Prediction for Deep Classifier via Label Ranking
Abstract
Conformal prediction is a statistical framework that generates prediction sets containing ground-truth labels with a desired coverage guarantee. The predicted probabilities produced by machine learning models are generally miscalibrated, leading to large prediction sets in conformal prediction. In this paper, we empirically and theoretically show that disregarding the probabilities' value will mitigate the undesirable effect of miscalibrated probability values. Then, we propose a novel algorithm named $\textit{Sorted Adaptive prediction sets}$ (SAPS), which discards all the probability values except for the maximum softmax probability. The key idea behind SAPS is to minimize the dependence of the non-conformity score on the probability values while retaining the uncertainty information. In this manner, SAPS can produce sets of small size and communicate instance-wise uncertainty. Theoretically, we provide a finite-sample coverage guarantee of SAPS and show that the expected value of set size from SAPS is always smaller than APS. Extensive experiments validate that SAPS not only lessens the prediction sets but also broadly enhances the conditional coverage rate and adaptation of prediction sets.
Solution for SMART-101 Challenge of ICCV Multi-modal Algorithmic Reasoning Task 2023
Authors: Xiangyu Wu, Yang Yang, Shengdong Xu, Yifeng Wu, Qingguo Chen, Jianfeng Lu
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
In this paper, we present our solution to a Multi-modal Algorithmic Reasoning Task: SMART-101 Challenge. Different from the traditional visual question-answering datasets, this challenge evaluates the abstraction, deduction, and generalization abilities of neural networks in solving visuolinguistic puzzles designed specifically for children in the 6-8 age group. We employed a divide-and-conquer approach. At the data level, inspired by the challenge paper, we categorized the whole questions into eight types and utilized the llama-2-chat model to directly generate the type for each question in a zero-shot manner. Additionally, we trained a yolov7 model on the icon45 dataset for object detection and combined it with the OCR method to recognize and locate objects and text within the images. At the model level, we utilized the BLIP-2 model and added eight adapters to the image encoder VIT-G to adaptively extract visual features for different question types. We fed the pre-constructed question templates as input and generated answers using the flan-t5-xxl decoder. Under the puzzle splits configuration, we achieved an accuracy score of 26.5 on the validation set and 24.30 on the private test set.
Query-dominant User Interest Network for Large-Scale Search Ranking
Abstract
Historical behaviors have shown great effect and potential in various prediction tasks, including recommendation and information retrieval. The overall historical behaviors are various but noisy while search behaviors are always sparse. Most existing approaches in personalized search ranking adopt the sparse search behaviors to learn representation with bottleneck, which do not sufficiently exploit the crucial long-term interest. In fact, there is no doubt that user long-term interest is various but noisy for instant search, and how to exploit it well still remains an open problem. To tackle this problem, in this work, we propose a novel model named Query-dominant user Interest Network (QIN), including two cascade units to filter the raw user behaviors and reweigh the behavior subsequences. Specifically, we propose a relevance search unit (RSU), which aims to search a subsequence relevant to the query first and then search the sub-subsequences relevant to the target item. These items are then fed into an attention unit called Fused Attention Unit (FAU). It should be able to calculate attention scores from the ID field and attribute field separately, and then adaptively fuse the item embedding and content embedding based on the user engagement of past period. Extensive experiments and ablation studies on real-world datasets demonstrate the superiority of our model over state-of-the-art methods. The QIN now has been successfully deployed on Kuaishou search, an online video search platform, and obtained 7.6% improvement on CTR.
Asynchronous Federated Learning with Incentive Mechanism Based on Contract Theory
Abstract
To address the challenges posed by the heterogeneity inherent in federated learning (FL) and to attract high-quality clients, various incentive mechanisms have been employed. However, existing incentive mechanisms are typically utilized in conventional synchronous aggregation, resulting in significant straggler issues. In this study, we propose a novel asynchronous FL framework that integrates an incentive mechanism based on contract theory. Within the incentive mechanism, we strive to maximize the utility of the task publisher by adaptively adjusting clients' local model training epochs, taking into account factors such as time delay and test accuracy. In the asynchronous scheme, considering client quality, we devise aggregation weights and an access control algorithm to facilitate asynchronous aggregation. Through experiments conducted on the MNIST dataset, the simulation results demonstrate that the test accuracy achieved by our framework is 3.12% and 5.84% higher than that achieved by FedAvg and FedProx without any attacks, respectively. The framework exhibits a 1.35% accuracy improvement over the ideal Local SGD under attacks. Furthermore, aiming for the same target accuracy, our framework demands notably less computation time than both FedAvg and FedProx.
Focus on Local Regions for Query-based Object Detection
Authors: Hongbin Xu, Yamei Xia, Shuai Zhao, Bo Cheng
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Query-based methods have garnered significant attention in object detection since the advent of DETR, the pioneering end-to-end query-based detector. However, these methods face challenges like slow convergence and suboptimal performance. Notably, self-attention in object detection often hampers convergence due to its global focus. To address these issues, we propose FoLR, a transformer-like architecture with only decoders. We enhance the self-attention mechanism by isolating connections between irrelevant objects that makes it focus on local regions but not global regions. We also design the adaptive sampling method to extract effective features based on queries' local regions from feature maps. Additionally, we employ a look-back strategy for decoders to retain prior information, followed by the Feature Mixer module to fuse features and queries. Experimental results demonstrate FoLR's state-of-the-art performance in query-based detectors, excelling in convergence speed and computational efficiency.
High-order adaptive multi-domain time integration scheme for microscale lithium-ion batteries simulations
Authors: Ali ASAD (CMAP), Romain de Loubens, Laurent François, Marc Massot (CMAP)
Abstract
We investigate the modeling and simulation of ionic transport and charge conservation in lithium-ion batteries (LIBs) at the microscale. It is a multiphysics problem that involves a wide range of time scales. The associated computational challenges motivate the investigation of numerical techniques that can decouple the time integration of the governing equations in the liquid electrolyte and the solid phase (active materials and current collectors). First, it is shown that semi-discretization in space of the non-dimensionalized governing equations leads to a system of index-1 semi-explicit differential algebraic equations (DAEs). Then, a new generation of strategies for multi-domain integration is presented, enabling high-order adaptive coupling of both domains in time. A simple 1D LIB half-cell code is implemented as a demonstrator of the new strategy for the simulation of different modes of cell operation. The integration of the decoupled subsystems is performed with high-order accurate implicit nonlinear solvers. The accuracy of the space discretization is assessed by comparing the numerical results to the analytical solutions. Then, temporal convergence studies demonstrate the accuracy of the new multi-domain coupling approach. Finally, the accuracy and computational efficiency of the adaptive coupling strategy are discussed in the light of the conditioning of the decoupled subproblems compared to the one of the fully-coupled problem. This new approach will constitute a key ingredient for the full scale 3D LIB high-fidelity simulations based on actual electrode microstructures.
Pi-DUAL: Using Privileged Information to Distinguish Clean from Noisy Labels
Authors: Ke Wang, Guillermo Ortiz-Jimenez, Rodolphe Jenatton, Mark Collier, Efi Kokiopoulou, Pascal Frossard
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Label noise is a pervasive problem in deep learning that often compromises the generalization performance of trained models. Recently, leveraging privileged information (PI) -- information available only during training but not at test time -- has emerged as an effective approach to mitigate this issue. Yet, existing PI-based methods have failed to consistently outperform their no-PI counterparts in terms of preventing overfitting to label noise. To address this deficiency, we introduce Pi-DUAL, an architecture designed to harness PI to distinguish clean from wrong labels. Pi-DUAL decomposes the output logits into a prediction term, based on conventional input features, and a noise-fitting term influenced solely by PI. A gating mechanism steered by PI adaptively shifts focus between these terms, allowing the model to implicitly separate the learning paths of clean and wrong labels. Empirically, Pi-DUAL achieves significant performance improvements on key PI benchmarks (e.g., +6.8% on ImageNet-PI), establishing a new state-of-the-art test set accuracy. Additionally, Pi-DUAL is a potent method for identifying noisy samples post-training, outperforming other strong methods at this task. Overall, Pi-DUAL is a simple, scalable and practical approach for mitigating the effects of label noise in a variety of real-world scenarios with PI.
EARL: Eye-on-Hand Reinforcement Learner for Dynamic Grasping with Active Pose Estimation
Abstract
In this paper, we explore the dynamic grasping of moving objects through active pose tracking and reinforcement learning for hand-eye coordination systems. Most existing vision-based robotic grasping methods implicitly assume target objects are stationary or moving predictably. Performing grasping of unpredictably moving objects presents a unique set of challenges. For example, a pre-computed robust grasp can become unreachable or unstable as the target object moves, and motion planning must also be adaptive. In this work, we present a new approach, Eye-on-hAnd Reinforcement Learner (EARL), for enabling coupled Eye-on-Hand (EoH) robotic manipulation systems to perform real-time active pose tracking and dynamic grasping of novel objects without explicit motion prediction. EARL readily addresses many thorny issues in automated hand-eye coordination, including fast-tracking of 6D object pose from vision, learning control policy for a robotic arm to track a moving object while keeping the object in the camera's field of view, and performing dynamic grasping. We demonstrate the effectiveness of our approach in extensive experiments validated on multiple commercial robotic arms in both simulations and complex real-world tasks.
Keyword: efficient
Deep Learning based Tomato Disease Detection and Remedy Suggestions using Mobile Application
Robust and Efficient Interference Neural Networks for Defending Against Adversarial Attacks in ImageNet
Reducing the False Positive Rate Using Bayesian Inference in Autonomous Driving Perception
DynamicBEV: Leveraging Dynamic Queries and Temporal Context for 3D Object Detection
LCOT: Linear circular optimal transport
LLM for SoC Security: A Paradigm Shift
Towards Agility: A Momentum Aware Trajectory Optimisation Framework using Full-Centroidal Dynamics & Implicit Inverse Kinematics
Numerical Analysis of time-dependent Hamilton-Jacobi Equations on Networks
QR-Tag: Angular Measurement and Tracking with a QR-Design Marker
When is Agnostic Reinforcement Learning Statistically Tractable?
On Time Domain Conformer Models for Monaural Speech Separation in Noisy Reverberant Acoustic Environments
Learning Layer-wise Equivariances Automatically using Gradients
Predicting Player Engagement in Tom Clancy's The Division 2: A Multimodal Approach via Pixels and Gamepad Actions
On the Correlation between Random Variables and their Principal Components
Entropy Based Multi-robot Active SLAM
CAW-coref: Conjunction-Aware Word-level Coreference Resolution
Look-Up mAI GeMM: Increasing AI GeMMs Performance by Nearly 2.5x via msGeMM
Automatic Integration for Spatiotemporal Neural Point Processes
Quasi-Monte Carlo sparse grid Galerkin finite element methods for linear elasticity equations with uncertainties
Motion Memory: Leveraging Past Experiences to Accelerate Future Motion Planning
CAT-RRT: Motion Planning that Admits Contact One Link at a Time
GeoLLM: Extracting Geospatial Knowledge from Large Language Models
CoT3DRef: Chain-of-Thoughts Data-Efficient 3D Visual Grounding
Spiking PointNet: Spiking Neural Networks for Point Clouds
Low-Rank Tensor Completion via Novel Sparsity-Inducing Regularizers
Efficient Adaptation of Large Vision Transformer via Adapter Re-Composing
Sample-Efficient Multi-Agent RL: An Optimization Perspective
A Unified View on Solving Objective Mismatch in Model-Based Reinforcement Learning
Bi-Level Offline Policy Optimization with Limited Exploration
Towards More Efficient Depression Risk Recognition via Gait
Dynamical versus Bayesian Phase Transitions in a Toy Model of Superposition
Learning bounded-degree polytrees with known skeleton
Boosting Continuous Control with Consistency Policy
Filter Pruning For CNN With Enhanced Linear Representation Redundancy
JointNet: Extending Text-to-Image Diffusion for Dense Distribution Modeling
Advanced Efficient Strategy for Detection of Dark Objects Based on Spiking Network with Multi-Box Detection
Rethinking Model Selection and Decoding for Keyphrase Generation with Pre-trained Sequence-to-Sequence Models
What Makes for Robust Multi-Modal Models in the Face of Missing Modalities?
Learning Stackable and Skippable LEGO Bricks for Efficient, Reconfigurable, and Variable-Resolution Diffusion Modeling
Top of the Heap: Efficient Memory Error Protection for Many Heap Objects
TANGO: Time-Reversal Latent GraphODE for Multi-Agent Dynamical Systems
Whispering LLaMA: A Cross-Modal Generative Error Correction Framework for Speech Recognition
MemSum-DQA: Adapting An Efficient Long Document Extractive Summarizer for Document Question Answering
Rule Mining for Correcting Classification Models
A Geometrical Approach to Evaluate the Adversarial Robustness of Deep Neural Networks
Topological data analysis of human vowels: Persistent homologies across representation spaces
Self-Supervised Set Representation Learning for Unsupervised Meta-Learning
Watt For What: Rethinking Deep Learning's Energy-Performance Relationship
Realizing Stabilized Landing for Computation-Limited Reusable Rockets: A Quantum Reinforcement Learning Approach
Efficient Retrieval of Images with Irregular Patterns using Morphological Image Analysis: Applications to Industrial and Healthcare datasets
Energy-Efficient Visual Search by Eye Movement and Low-Latency Spiking Neural Network
FTFT: efficient and robust Fine-Tuning by transFerring Training dynamics
EViT: An Eagle Vision Transformer with Bi-Fovea Self-Attention
A Survey on Intelligent Iterative Methods for Solving Sparse Linear Algebraic Equations
The Lattice Overparametrization Paradigm for the Machine Learning of Lattice Operators
Self-Supervised Representation Learning for Online Handwriting Text Classification
A Parallelized, Adam-Based Solver for Reserve and Security Constrained AC Unit Commitment
Unlock the Potential of Counterfactually-Augmented Data in Out-Of-Distribution Generalization
Machine Learning Quantum Systems with Magnetic p-bits
DeepLSH: Deep Locality-Sensitive Hash Learning for Fast and Efficient Near-Duplicate Crash Report Detection
Comparing AI Algorithms for Optimizing Elliptic Curve Cryptography Parameters in Third-Party E-Commerce Integrations: A Pre-Quantum Era Analysis
Going Beyond Neural Network Feature Similarity: The Network Feature Complexity and Its Interpretation Using Category Theory
Correlated Noise Provably Beats Independent Noise for Differentially Private Learning
Uni3D: Exploring Unified 3D Representation at Scale
Information Content Exploration
A Supervised Embedding and Clustering Anomaly Detection method for classification of Mobile Network Faults
Spectral Entry-wise Matrix Estimation for Low-Rank Reinforcement Learning
$f$-Policy Gradients: A General Framework for Goal Conditioned RL using $f$-Divergences
Inverse Factorized Q-Learning for Cooperative Multi-agent Imitation Learning
Teaching Language Models to Hallucinate Less with Synthetic Tasks
Keyword: faster
Analysis of Learned Features and Framework for Potato Disease Detection
DockGame: Cooperative Games for Multimeric Rigid Protein Docking
Words into Action: Learning Diverse Humanoid Robot Behaviors using Language Guided Iterative Motion Refinement
SCAR: Power Side-Channel Analysis at RTL-Level
Exploring Users Pointing Performance on Large Displays with Different Curvatures in Virtual Reality
Just-in-Time Flaky Test Detection via Abstracted Failure Symptom Matching
Deep Learning for Automatic Detection and Facial Recognition in Japanese Macaques: Illuminating Social Networks
Characterization of the Complexity of Computing the Capacity of Colored Noise Gaussian Channels
Hierarchical Mask2Former: Panoptic Segmentation of Crops, Weeds and Leaves
FTFT: efficient and robust Fine-Tuning by transFerring Training dynamics
EViT: An Eagle Vision Transformer with Bi-Fovea Self-Attention
Mistral 7B
Keyword: mobile
Deep Learning based Tomato Disease Detection and Remedy Suggestions using Mobile Application
Review of control algorithms for mobile robotics
Layout Sequence Prediction From Noisy Mobile Modality
High-Fidelity 3D Head Avatars Reconstruction through Spatially-Varying Expression Conditioned Neural Radiance Field
MEC-Intelligent Agent Support for Low-Latency Data Plane in Private NextG Core
Redundant and Loosely Coupled LiDAR-Wi-Fi Integration for Robust Global Localization in Autonomous Mobile Robotics
A Supervised Embedding and Clustering Anomaly Detection method for classification of Mobile Network Faults
Keyword: pruning
Compressing Context to Enhance Inference Efficiency of Large Language Models
SUBP: Soft Uniform Block Pruning for 1xN Sparse CNNs Multithreading Acceleration
Filter Pruning For CNN With Enhanced Linear Representation Redundancy
Skeleton Ground Truth Extraction: Methodology, Annotation Tool and Benchmarks
Rule Mining for Correcting Classification Models
Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning
Keyword: diffusion
DF-3DFace: One-to-Many Speech Synchronized 3D Face Animation with Diffusion
Stabilized finite elements for the solution of the Reynolds equation considering cavitation
Layout Sequence Prediction From Noisy Mobile Modality
Latent Diffusion Model for DNA Sequence Generation
Memory-Consistent Neural Networks for Imitation Learning
behavior cloning'' for conveniently training the policy from nothing more than pre-recorded demonstrations, but carefully design the model class to counter the compounding error phenomenon. Our
memory-consistent neural network'' (MCNN) outputs are hard-constrained to stay within clearly specified permissible regions anchored to prototypical ``memory'' training samples. We provide a guaranteed upper bound for the sub-optimality gap induced by MCNN policies. Using MCNNs on 9 imitation learning tasks, with MLP, Transformer, and Diffusion backbones, spanning dexterous robotic manipulation and driving, proprioceptive inputs and visual inputs, and varying sizes and types of demonstration data, we find large and consistent gains in performance, validating that MCNNs are better-suited than vanilla deep neural networks for imitation learning applications. Website: https://sites.google.com/view/mcnn-imitationDockGame: Cooperative Games for Multimeric Rigid Protein Docking
Improving Compositional Text-to-image Generation with Large Vision-Language Models
Advancing Pose-Guided Image Synthesis with Progressive Conditional Diffusion Models
Boosting Continuous Control with Consistency Policy
JointNet: Extending Text-to-Image Diffusion for Dense Distribution Modeling
Leveraging Diffusion-Based Image Variations for Robust Training on Poisoned Data
Learning Stackable and Skippable LEGO Bricks for Efficient, Reconfigurable, and Variable-Resolution Diffusion Modeling
Advective Diffusion Transformers for Topological Generalization in Graph Learning
AnoDODE: Anomaly Detection with Diffusion ODE
Tertiary Lymphoid Structures Generation through Graph-based Diffusion
Latent Diffusion Counterfactual Explanations
HiFi-123: Towards High-fidelity One Image to 3D Content Generation
What Does Stable Diffusion Know about the 3D Scene?
Keyword: adaptive
Enhancing Document-level Event Argument Extraction with Contextual Clues and Role Relevance
A Natural Indirect Adaptive Controller for a Satellite-Mounted Manipulator
CAT-RRT: Motion Planning that Admits Contact One Link at a Time
Efficient Adaptation of Large Vision Transformer via Adapter Re-Composing
A Unified View on Solving Objective Mismatch in Model-Based Reinforcement Learning
High-Fidelity 3D Head Avatars Reconstruction through Spatially-Varying Expression Conditioned Neural Radiance Field
Conformal Prediction for Deep Classifier via Label Ranking
Solution for SMART-101 Challenge of ICCV Multi-modal Algorithmic Reasoning Task 2023
Query-dominant User Interest Network for Large-Scale Search Ranking
Asynchronous Federated Learning with Incentive Mechanism Based on Contract Theory
Focus on Local Regions for Query-based Object Detection
High-order adaptive multi-domain time integration scheme for microscale lithium-ion batteries simulations
Pi-DUAL: Using Privileged Information to Distinguish Clean from Noisy Labels
EARL: Eye-on-Hand Reinforcement Learner for Dynamic Grasping with Active Pose Estimation
Keyword: quantization
There is no result