Abstract
This project introduces a groundbreaking approach to address the challenge of periodic signal compression. By proposing a novel adaptive coding method, coupled with hardware-assisted data compression, we have developed a new architecture model tailored for efficient data compression. The selected compression scheme has demonstrated remarkable results, showcasing reduced memory communication volume and power consumption in the cache memory path of benchmark systems. With a reduction range of 4.2% to 35.2%, this innovation paves the way for affordable smart sensing, monitoring, diagnostics, and protection in emerging low-cost device types. Consequently, this cutting-edge technology enhances electrical signal compression and contributes to grid improvement. Additionally, we explore the novel application of harnessing wasted thermal energy in the Read-Only Memory (ROM) using thermoelectricity (TE). This approach captures the excess thermal energy, converting it into electrical energy through optimized supercapacitor charging, resulting in efficient energy utilization. This innovation intersects the fields of embedded systems, data compression, energy efficiency, and smart grid technology.
Pseudo-online framework for BCI evaluation: A MOABB perspective
Authors: Igor Carrara (UCA, CRISAM), Théodore Papadopoulo (UCA, CRISAM)
Abstract
Objective: BCI (Brain-Computer Interface) technology operates in three modes: online, offline, and pseudo-online. In the online mode, real-time EEG data is constantly analyzed. In offline mode, the signal is acquired and processed afterwards. The pseudo-online mode processes collected data as if they were received in real-time. The main difference is that the offline mode often analyzes the whole data, while the online and pseudo-online modes only analyze data in short time windows. Offline analysis is usually done with asynchronous BCIs, which restricts analysis to predefined time windows. Asynchronous BCI, compatible with online and pseudo-online modes, allows flexible mental activity duration. Offline processing tends to be more accurate, while online analysis is better for therapeutic applications. Pseudo-online implementation approximates online processing without real-time constraints. Many BCI studies being offline introduce biases compared to real-life scenarios, impacting classification algorithm performance. Approach: The objective of this research paper is therefore to extend the current MOABB framework, operating in offline mode, so as to allow a comparison of different algorithms in a pseudo-online setting with the use of a technology based on overlapping sliding windows. To do this will require the introduction of a idle state event in the dataset that takes into account all different possibilities that are not task thinking. To validate the performance of the algorithms we will use the normalized Matthews Correlation Coefficient (nMCC) and the Information Transfer Rate (ITR). Main results: We analyzed the state-of-the-art algorithms of the last 15 years over several Motor Imagery (MI) datasets composed by several subjects, showing the differences between the two approaches from a statistical point of view. Significance: The ability to analyze the performance of different algorithms in offline and pseudo-online modes will allow the BCI community to obtain more accurate and comprehensive reports regarding the performance of classification algorithms.
An engine to simulate insurance fraud network data
Abstract
Traditionally, the detection of fraudulent insurance claims relies on business rules and expert judgement which makes it a time-consuming and expensive process (\'Oskarsd\'ottir et al., 2022). Consequently, researchers have been examining ways to develop efficient and accurate analytic strategies to flag suspicious claims. Feeding learning methods with features engineered from the social network of parties involved in a claim is a particularly promising strategy (see for example Van Vlasselaer et al. (2016); Tumminello et al. (2023)). When developing a fraud detection model, however, we are confronted with several challenges. The uncommon nature of fraud, for example, creates a high class imbalance which complicates the development of well performing analytic classification models. In addition, only a small number of claims are investigated and get a label, which results in a large corpus of unlabeled data. Yet another challenge is the lack of publicly available data. This hinders not only the development of new methods, but also the validation of existing techniques. We therefore design a simulation machine that is engineered to create synthetic data with a network structure and available covariates similar to the real life insurance fraud data set analyzed in \'Oskarsd\'ottir et al. (2022). Further, the user has control over several data-generating mechanisms. We can specify the total number of policyholders and parties, the desired level of imbalance and the (effect size of the) features in the fraud generating model. As such, the simulation engine enables researchers and practitioners to examine several methodological challenges as well as to test their (development strategy of) insurance fraud detection models in a range of different settings. Moreover, large synthetic data sets can be generated to evaluate the predictive performance of (advanced) machine learning techniques.
VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection
Abstract
The recent contrastive language-image pre-training (CLIP) model has shown great success in a wide range of image-level tasks, revealing remarkable ability for learning powerful visual representations with rich semantics. An open and worthwhile problem is efficiently adapting such a strong model to the video domain and designing a robust video anomaly detector. In this work, we propose VadCLIP, a new paradigm for weakly supervised video anomaly detection (WSVAD) by leveraging the frozen CLIP model directly without any pre-training and fine-tuning process. Unlike current works that directly feed extracted features into the weakly supervised classifier for frame-level binary classification, VadCLIP makes full use of fine-grained associations between vision and language on the strength of CLIP and involves dual branch. One branch simply utilizes visual features for coarse-grained binary classification, while the other fully leverages the fine-grained language-image alignment. With the benefit of dual branch, VadCLIP achieves both coarse-grained and fine-grained video anomaly detection by transferring pre-trained knowledge from CLIP to WSVAD task. We conduct extensive experiments on two commonly-used benchmarks, demonstrating that VadCLIP achieves the best performance on both coarse-grained and fine-grained WSVAD, surpassing the state-of-the-art methods by a large margin. Specifically, VadCLIP achieves 84.51% AP and 88.02% AUC on XD-Violence and UCF-Crime, respectively. Code and features will be released to facilitate future VAD research.
Abstract
The increasing versatility of language models LMs has given rise to a new class of benchmarks that comprehensively assess a broad range of capabilities. Such benchmarks are associated with massive computational costs reaching thousands of GPU hours per model. However the efficiency aspect of these evaluation efforts had raised little discussion in the literature. In this work we present the problem of Efficient Benchmarking namely intelligently reducing the computation costs of LM evaluation without compromising reliability. Using the HELM benchmark as a test case we investigate how different benchmark design choices affect the computation-reliability tradeoff. We propose to evaluate the reliability of such decisions by using a new measure Decision Impact on Reliability DIoR for short. We find for example that the current leader on HELM may change by merely removing a low-ranked model from the benchmark and observe that a handful of examples suffice to obtain the correct benchmark ranking. Conversely a slightly different choice of HELM scenarios varies ranking widely. Based on our findings we outline a set of concrete recommendations for more efficient benchmark design and utilization practices leading to dramatic cost savings with minimal loss of benchmark reliability often reducing computation by x100 or more.
Abstract
This paper presents VBMO, the Voting-Based Multi-Objective path planning algorithm, that generates optimal single-objective plans, evaluates each of them with respect to the other objectives, and selects one with a voting mechanism. VBMO does not use hand-tuned weights, consider the multiple objectives at every step of search, or use an evolutionary algorithm. Instead, it considers how a plan that is optimal in one objective may perform well with respect to others. VBMO incorporates three voting mechanisms: range, Borda, and combined approval. Extensive evaluation in diverse and complex environments demonstrates the algorithm's ability to efficiently produce plans that satisfy multiple objectives.
Weakly Supervised Face and Whole Body Recognition in Turbulent Environments
Authors: Kshitij Nikhal, Benjamin S. Riggan
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Face and person recognition have recently achieved remarkable success under challenging scenarios, such as off-pose and cross-spectrum matching. However, long-range recognition systems are often hindered by atmospheric turbulence, leading to spatially and temporally varying distortions in the image. Current solutions rely on generative models to reconstruct a turbulent-free image, but often preserve photo-realism instead of discriminative features that are essential for recognition. This can be attributed to the lack of large-scale datasets of turbulent and pristine paired images, necessary for optimal reconstruction. To address this issue, we propose a new weakly supervised framework that employs a parameter-efficient self-attention module to generate domain agnostic representations, aligning turbulent and pristine images into a common subspace. Additionally, we introduce a new tilt map estimator that predicts geometric distortions observed in turbulent images. This estimate is used to re-rank gallery matches, resulting in up to 13.86\% improvement in rank-1 accuracy. Our method does not require synthesizing turbulent-free images or ground-truth paired images, and requires significantly fewer annotated samples, enabling more practical and rapid utility of increasingly large datasets. We analyze our framework using two datasets -- Long-Range Face Identification Dataset (LRFID) and BRIAR Government Collection 1 (BGC1) -- achieving enhanced discriminability under varying turbulence and standoff distance.
3ET: Efficient Event-based Eye Tracking using a Change-Based ConvLSTM Network
Abstract
This paper presents a sparse Change-Based Convolutional Long Short-Term Memory (CB-ConvLSTM) model for event-based eye tracking, key for next-generation wearable healthcare technology such as AR/VR headsets. We leverage the benefits of retina-inspired event cameras, namely their low-latency response and sparse output event stream, over traditional frame-based cameras. Our CB-ConvLSTM architecture efficiently extracts spatio-temporal features for pupil tracking from the event stream, outperforming conventional CNN structures. Utilizing a delta-encoded recurrent path enhancing activation sparsity, CB-ConvLSTM reduces arithmetic operations by approximately 4.7$\times$ without losing accuracy when tested on a \texttt{v2e}-generated event dataset of labeled pupils. This increase in efficiency makes it ideal for real-time eye tracking in resource-constrained devices. The project code and dataset are openly available at \url{https://github.com/qinche106/cb-convlstm-eyetracking}.
Understanding Hessian Alignment for Domain Generalization
Authors: Sobhan Hemati, Guojun Zhang, Amir Estiri, Xi Chen
Abstract
Out-of-distribution (OOD) generalization is a critical ability for deep learning models in many real-world scenarios including healthcare and autonomous vehicles. Recently, different techniques have been proposed to improve OOD generalization. Among these methods, gradient-based regularizers have shown promising performance compared with other competitors. Despite this success, our understanding of the role of Hessian and gradient alignment in domain generalization is still limited. To address this shortcoming, we analyze the role of the classifier's head Hessian matrix and gradient in domain generalization using recent OOD theory of transferability. Theoretically, we show that spectral norm between the classifier's head Hessian matrices across domains is an upper bound of the transfer measure, a notion of distance between target and source domains. Furthermore, we analyze all the attributes that get aligned when we encourage similarity between Hessians and gradients. Our analysis explains the success of many regularizers like CORAL, IRM, V-REx, Fish, IGA, and Fishr as they regularize part of the classifier's head Hessian and/or gradient. Finally, we propose two simple yet effective methods to match the classifier's head Hessians and gradients in an efficient way, based on the Hessian Gradient Product (HGP) and Hutchinson's method (Hutchinson), and without directly calculating Hessians. We validate the OOD generalization ability of proposed methods in different scenarios, including transferability, severe correlation shift, label shift and diversity shift. Our results show that Hessian alignment methods achieve promising performance on various OOD benchmarks. The code is available at \url{https://github.com/huawei-noah/Federated-Learning/tree/main/HessianAlignment}.
Few-shot Anomaly Detection in Text with Deviation Learning
Abstract
Most current methods for detecting anomalies in text concentrate on constructing models solely relying on unlabeled data. These models operate on the presumption that no labeled anomalous examples are available, which prevents them from utilizing prior knowledge of anomalies that are typically present in small numbers in many real-world applications. Furthermore, these models prioritize learning feature embeddings rather than optimizing anomaly scores directly, which could lead to suboptimal anomaly scoring and inefficient use of data during the learning process. In this paper, we introduce FATE, a deep few-shot learning-based framework that leverages limited anomaly examples and learns anomaly scores explicitly in an end-to-end method using deviation learning. In this approach, the anomaly scores of normal examples are adjusted to closely resemble reference scores obtained from a prior distribution. Conversely, anomaly samples are forced to have anomalous scores that considerably deviate from the reference score in the upper tail of the prior. Additionally, our model is optimized to learn the distinct behavior of anomalies by utilizing a multi-head self-attention layer and multiple instance learning approaches. Comprehensive experiments on several benchmark datasets demonstrate that our proposed approach attains a new level of state-of-the-art performance.
Resource Allocation in Cloud Computing Using Genetic Algorithm and Neural Network
Abstract
Cloud computing is one of the most used distributed systems for data processing and data storage. Due to the continuous increase in the size of the data processed by cloud computing, scheduling multiple tasks to maintain efficiency while reducing idle becomes more and more challenging. Efficient cloud-based scheduling is also highly sought by modern transportation systems to improve their security. In this paper, we propose a hybrid algorithm that leverages genetic algorithms and neural networks to improve scheduling. Our method classifies tasks with the Neural Network Task Classification (N2TC) and sends the selected tasks to the Genetic Algorithm Task Assignment (GATA) to allocate resources. It is fairness aware to prevent starvation and considers the execution time, response time, cost, and system efficiency. Evaluations show that our approach outperforms the state-of-the-art method by 3.2% at execution time, 13.3% in costs, and 12.1% at response time.
Towards Safe Automated Refactoring of Imperative Deep Learning Programs to Graph Execution
Authors: Raffi Khatchadourian, Tatiana Castro Vélez, Mehdi Bagherzadeh, Nan Jia, Anita Raja
Subjects: Software Engineering (cs.SE); Programming Languages (cs.PL)
Abstract
Efficiency is essential to support responsiveness w.r.t. ever-growing datasets, especially for Deep Learning (DL) systems. DL frameworks have traditionally embraced deferred execution-style DL code -- supporting symbolic, graph-based Deep Neural Network (DNN) computation. While scalable, such development tends to produce code that is error-prone, non-intuitive, and difficult to debug. Consequently, more natural, less error-prone imperative DL frameworks encouraging eager execution have emerged at the expense of run-time performance. Though hybrid approaches aim for the "best of both worlds," using them effectively requires subtle considerations to make code amenable to safe, accurate, and efficient graph execution -- avoiding performance bottlenecks and semantically inequivalent results. We present our ongoing work on an automated refactoring approach that assists developers in specifying whether and how their otherwise eagerly-executed imperative DL code could be reliably and efficiently executed as graphs at run-time in a semantics-preserving fashion. The approach, based on a novel tensor analysis specifically for imperative DL code, consists of refactoring preconditions for automatically determining when it is safe and potentially advantageous to migrate imperative DL code to graph execution and modifying decorator parameters or eagerly executing code already running as graphs. The approach is being implemented as a PyDev Eclipse IDE plug-in and uses the WALA Ariadne analysis framework. We discuss our ongoing work towards optimizing imperative DL code to its full potential.
HypBO: Expert-Guided Chemist-in-the-Loop Bayesian Search for New Materials
Authors: Abdoulatif Cisse, Xenophon Evangelopoulos, Sam Carruthers, Vladimir V. Gusev, Andrew I. Cooper
Abstract
Robotics and automation offer massive accelerations for solving intractable, multivariate scientific problems such as materials discovery, but the available search spaces can be dauntingly large. Bayesian optimization (BO) has emerged as a popular sample-efficient optimization engine, thriving in tasks where no analytic form of the target function/property is known. Here we exploit expert human knowledge in the form of hypotheses to direct Bayesian searches more quickly to promising regions of chemical space. Previous methods have used underlying distributions derived from existing experimental measurements, which is unfeasible for new, unexplored scientific tasks. Also, such distributions cannot capture intricate hypotheses. Our proposed method, which we call HypBO, uses expert human hypotheses to generate an improved seed of samples. Unpromising seeds are automatically discounted, while promising seeds are used to augment the surrogate model data, thus achieving better-informed sampling. This process continues in a global versus local search fashion, organized in a bilevel optimization framework. We validate the performance of our method on a range of synthetic functions and demonstrate its practical utility on a real chemical design task where the use of expert hypotheses accelerates the search performance significantly.
Karasu: A Collaborative Approach to Efficient Cluster Configuration for Big Data Analytics
Authors: Dominik Scheinert, Philipp Wiesner, Thorsten Wittkopp, Lauritz Thamsen, Jonathan Will, Odej Kao
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
Abstract
Selecting the right resources for big data analytics jobs is hard because of the wide variety of configuration options like machine type and cluster size. As poor choices can have a significant impact on resource efficiency, cost, and energy usage, automated approaches are gaining popularity. Most existing methods rely on profiling recurring workloads to find near-optimal solutions over time. Due to the cold-start problem, this often leads to lengthy and costly profiling phases. However, big data analytics jobs across users can share many common properties: they often operate on similar infrastructure, using similar algorithms implemented in similar frameworks. The potential in sharing aggregated profiling runs to collaboratively address the cold start problem is largely unexplored. We present Karasu, an approach to more efficient resource configuration profiling that promotes data sharing among users working with similar infrastructures, frameworks, algorithms, or datasets. Karasu trains lightweight performance models using aggregated runtime information of collaborators and combines them into an ensemble method to exploit inherent knowledge of the configuration search space. Moreover, Karasu allows the optimization of multiple objectives simultaneously. Our evaluation is based on performance data from diverse workload executions in a public cloud environment. We show that Karasu is able to significantly boost existing methods in terms of performance, search time, and cost, even when few comparable profiling runs are available that share only partial common characteristics with the target job.
Flexible Multi-DoF Aerial 3D Printing Supported with Automated Optimal Chunking
Authors: Marios-Nektarios Stamatopoulos, Avijit Banerjee, George Nikolakopoulos
Abstract
The future of 3D printing utilizing unmanned aerial vehicles (UAVs) presents a promising capability to revolutionize manufacturing and to enable the creation of large-scale structures in remote and hard- to-reach areas e.g. in other planetary systems. Nevertheless, the limited payload capacity of UAVs and the complexity in the 3D printing of large objects pose significant challenges. In this article we propose a novel chunk-based framework for distributed 3D printing using UAVs that sets the basis for a fully collaborative aerial 3D printing of challenging structures. The presented framework, through a novel proposed optimisation process, is able to divide the 3D model to be printed into small, manageable chunks and to assign them to a UAV for partial printing of the assigned chunk, in a fully autonomous approach. Thus, we establish the algorithms for chunk division, allocation, and printing, and we also introduce a novel algorithm that efficiently partitions the mesh into planar chunks, while accounting for the inter-connectivity constraints of the chunks. The efficiency of the proposed framework is demonstrated through multiple physics based simulations in Gazebo, where a CAD construction mesh is printed via multiple UAVs carrying materials whose volume is proportionate to a fraction of the total mesh volume.
Integrated Image and Location Analysis for Wound Classification: A Deep Learning Approach
Abstract
The global burden of acute and chronic wounds presents a compelling case for enhancing wound classification methods, a vital step in diagnosing and determining optimal treatments. Recognizing this need, we introduce an innovative multi-modal network based on a deep convolutional neural network for categorizing wounds into four categories: diabetic, pressure, surgical, and venous ulcers. Our multi-modal network uses wound images and their corresponding body locations for more precise classification. A unique aspect of our methodology is incorporating a body map system that facilitates accurate wound location tagging, improving upon traditional wound image classification techniques. A distinctive feature of our approach is the integration of models such as VGG16, ResNet152, and EfficientNet within a novel architecture. This architecture includes elements like spatial and channel-wise Squeeze-and-Excitation modules, Axial Attention, and an Adaptive Gated Multi-Layer Perceptron, providing a robust foundation for classification. Our multi-modal network was trained and evaluated on two distinct datasets comprising relevant images and corresponding location information. Notably, our proposed network outperformed traditional methods, reaching an accuracy range of 74.79% to 100% for Region of Interest (ROI) without location classifications, 73.98% to 100% for ROI with location classifications, and 78.10% to 100% for whole image classifications. This marks a significant enhancement over previously reported performance metrics in the literature. Our results indicate the potential of our multi-modal network as an effective decision-support tool for wound image classification, paving the way for its application in various clinical contexts.
Cabrita: closing the gap for foreign languages
Authors: Celio Larcher, Marcos Piau, Paulo Finardi, Pedro Gengo, Piero Esposito, Vinicius Caridá
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Abstract
The strategy of training the model from scratch in a specific language or domain serves two essential purposes: i) enhancing performance in the particular linguistic or domain context, and ii) ensuring effective tokenization. The main limitation inherent to this approach lies in the associated cost, which can reach six to seven-digit dollar values, depending on the model size and the number of parameters involved. The main solution to overcome the cost challenge is to rely on available pre-trained models, which, despite recent advancements such as the LLaMA and LLaMA-2 models, still demonstrate inefficiency for certain specific domain problems or prove ineffective in scenarios involving conversational memory resources, given the large number of tokens required to represent text. To overcome this issue, we present a methodology named Cabrita, which, as our research demonstrates, successfully addresses the performance and efficient tokenization problem, all at an affordable cost. We believe that this methodology can be applied to any transformer-like architecture model. To validate the study, we conducted continuous pre-training exclusively using Portuguese text on a 3-billion-parameter model known as OpenLLaMA, resulting in a model named openCabrita 3B. The openCabrita 3B also features a new tokenizer that results in a significant reduction in the number of tokens required to represent the text. In our assessment, for few-shot learning tasks, we achieved similar results with this 3B model compared to a traditional continuous pre-training approach as well as to 7B models English pre-trained models.
The macroscopic finite-difference scheme and modified equations of the general propagation multiple-relaxation-time lattice Boltzmann model
Authors: Ying Chen, Xi Liu, Zhenhua Chai, Baochang Shi
Subjects: Numerical Analysis (math.NA); Analysis of PDEs (math.AP)
Abstract
In this paper, we first present the general propagation multiple-relaxation-time lattice Boltzmann (GPMRT-LB) model and obtain the corresponding macroscopic finite-difference (GPMFD) scheme on conservative moments. Then based on the Maxwell iteration method, we conduct the analysis on the truncation errors and modified equations (MEs) of the GPMRT-LB model and GPMFD scheme at both diffusive and acoustic scalings. For the nonlinear anisotropic convection-diffusion equation (NACDE) and Navier-Stokes equations (NSEs), we also derive the first- and second-order MEs of the GPMRT-LB model and GPMFD scheme. In particular, for the one-dimensional convection-diffusion equation (CDE) with the constant velocity and diffusion coefficient, we can develop a fourth-order GPMRT-LB (F-GPMRT-LB) model and the corresponding fourth-order GPMFD (F-GPMFD) scheme at the diffusive scaling. Finally, two benchmark problems, Gauss hill problem and Poiseuille flow in two-dimensional space, are used to test the GPMRT-LB model and GPMFD scheme, and it is found that the numerical results are not only in good agreement with corresponding analytical solutions, but also have a second-order convergence rate in space. Additionally, a numerical study on one-dimensional CDE also demonstrates that the F-GPMRT-LB model and F-GPMFD scheme can achieve a fourth-order accuracy in space, which is consistent with our theoretical analysis.
Addressing Selection Bias in Computerized Adaptive Testing: A User-Wise Aggregate Influence Function Approach
Authors: Soonwoo Kwon, Sojung Kim, Seunghyun Lee, Jin-Young Kim, Suyeong An, Kyuseok Kim
Subjects: Machine Learning (cs.LG); Computers and Society (cs.CY)
Abstract
Computerized Adaptive Testing (CAT) is a widely used, efficient test mode that adapts to the examinee's proficiency level in the test domain. CAT requires pre-trained item profiles, for CAT iteratively assesses the student real-time based on the registered items' profiles, and selects the next item to administer using candidate items' profiles. However, obtaining such item profiles is a costly process that involves gathering a large, dense item-response data, then training a diagnostic model on the collected data. In this paper, we explore the possibility of leveraging response data collected in the CAT service. We first show that this poses a unique challenge due to the inherent selection bias introduced by CAT, i.e., more proficient students will receive harder questions. Indeed, when naively training the diagnostic model using CAT response data, we observe that item profiles deviate significantly from the ground-truth. To tackle the selection bias issue, we propose the user-wise aggregate influence function method. Our intuition is to filter out users whose response data is heavily biased in an aggregate manner, as judged by how much perturbation the added data will introduce during parameter estimation. This way, we may enhance the performance of CAT while introducing minimal bias to the item profiles. We provide extensive experiments to demonstrate the superiority of our proposed method based on the three public datasets and one dataset that contains real-world CAT response data.
Adaptive Encoding Strategies for Erasing-Based Lossless Floating-Point Compression
Abstract
Lossless floating-point time series compression is crucial for a wide range of critical scenarios. Nevertheless, it is a big challenge to compress time series losslessly due to the complex underlying layouts of floating-point values. The state-of-the-art erasing-based compression algorithm Elf demonstrates a rather impressive performance. We give an in-depth exploration of the encoding strategies of Elf, and find that there is still much room for improvement. In this paper, we propose Elf, which employs a set of optimizations for leading zeros, center bits and sharing condition. Specifically, we develop a dynamic programming algorithm with a set of pruning strategies to compute the adaptive approximation rules efficiently. We theoretically prove that the adaptive approximation rules are globally optimal. We further extend Elf to Streaming Elf, i.e., SElf, which achieves almost the same compression ratio as Elf, while enjoying even higher efficiency in streaming scenarios. We compare Elf and SElf with 8 competitors using 22 datasets. The results demonstrate that SElf achieves 9.2% relative compression ratio improvement over the best streaming competitor while maintaining similar efficiency, and that Elf* ranks among the most competitive batch compressors. All source codes are publicly released.
LFS-GAN: Lifelong Few-Shot Image Generation
Authors: Juwon Seo, Ji-Su Kang, Gyeong-Moon Park
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
We address a challenging lifelong few-shot image generation task for the first time. In this situation, a generative model learns a sequence of tasks using only a few samples per task. Consequently, the learned model encounters both catastrophic forgetting and overfitting problems at a time. Existing studies on lifelong GANs have proposed modulation-based methods to prevent catastrophic forgetting. However, they require considerable additional parameters and cannot generate high-fidelity and diverse images from limited data. On the other hand, the existing few-shot GANs suffer from severe catastrophic forgetting when learning multiple tasks. To alleviate these issues, we propose a framework called Lifelong Few-Shot GAN (LFS-GAN) that can generate high-quality and diverse images in lifelong few-shot image generation task. Our proposed framework learns each task using an efficient task-specific modulator - Learnable Factorized Tensor (LeFT). LeFT is rank-constrained and has a rich representation ability due to its unique reconstruction technique. Furthermore, we propose a novel mode seeking loss to improve the diversity of our model in low-data circumstances. Extensive experiments demonstrate that the proposed LFS-GAN can generate high-fidelity and diverse images without any forgetting and mode collapse in various domains, achieving state-of-the-art in lifelong few-shot image generation task. Surprisingly, we find that our LFS-GAN even outperforms the existing few-shot GANs in the few-shot image generation task. The code is available at Github.
Diverse Policies Converge in Reward-free Markov Decision Processe
Abstract
Reinforcement learning has achieved great success in many decision-making tasks, and traditional reinforcement learning algorithms are mainly designed for obtaining a single optimal solution. However, recent works show the importance of developing diverse policies, which makes it an emerging research topic. Despite the variety of diversity reinforcement learning algorithms that have emerged, none of them theoretically answer the question of how the algorithm converges and how efficient the algorithm is. In this paper, we provide a unified diversity reinforcement learning framework and investigate the convergence of training diverse policies. Under such a framework, we also propose a provably efficient diversity reinforcement learning algorithm. Finally, we verify the effectiveness of our method through numerical experiments.
OFVL-MS: Once for Visual Localization across Multiple Indoor Scenes
Authors: Tao Xie, Kun Dai, Siyi Lu, Ke Wang, Zhiqiang Jiang, Jinghan Gao, Dedong Liu, Jie Xu, Lijun Zhao, Ruifeng Li
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
In this work, we seek to predict camera poses across scenes with a multi-task learning manner, where we view the localization of each scene as a new task. We propose OFVL-MS, a unified framework that dispenses with the traditional practice of training a model for each individual scene and relieves gradient conflict induced by optimizing multiple scenes collectively, enabling efficient storage yet precise visual localization for all scenes. Technically, in the forward pass of OFVL-MS, we design a layer-adaptive sharing policy with a learnable score for each layer to automatically determine whether the layer is shared or not. Such sharing policy empowers us to acquire task-shared parameters for a reduction of storage cost and task-specific parameters for learning scene-related features to alleviate gradient conflict. In the backward pass of OFVL-MS, we introduce a gradient normalization algorithm that homogenizes the gradient magnitude of the task-shared parameters so that all tasks converge at the same pace. Furthermore, a sparse penalty loss is applied on the learnable scores to facilitate parameter sharing for all tasks without performance degradation. We conduct comprehensive experiments on multiple benchmarks and our new released indoor dataset LIVL, showing that OFVL-MS families significantly outperform the state-of-the-arts with fewer parameters. We also verify that OFVL-MS can generalize to a new scene with much few parameters while gaining superior localization performance.
System Identification for Continuous-time Linear Dynamical Systems
Authors: Peter Halmos, Jonathan Pillow, David A. Knowles
Subjects: Machine Learning (cs.LG); Systems and Control (eess.SY)
Abstract
The problem of system identification for the Kalman filter, relying on the expectation-maximization (EM) procedure to learn the underlying parameters of a dynamical system, has largely been studied assuming that observations are sampled at equally-spaced time points. However, in many applications this is a restrictive and unrealistic assumption. This paper addresses system identification for the continuous-discrete filter, with the aim of generalizing learning for the Kalman filter by relying on a solution to a continuous-time It\^o stochastic differential equation (SDE) for the latent state and covariance dynamics. We introduce a novel two-filter, analytical form for the posterior with a Bayesian derivation, which yields analytical updates which do not require the forward-pass to be pre-computed. Using this analytical and efficient computation of the posterior, we provide an EM procedure which estimates the parameters of the SDE, naturally incorporating irregularly sampled measurements. Generalizing the learning of latent linear dynamical systems (LDS) to continuous-time may extend the use of the hybrid Kalman filter to data which is not regularly sampled or has intermittent missing values, and can extend the power of non-linear system identification methods such as switching LDS (SLDS), which rely on EM for the linear discrete-time Kalman filter as a sub-unit for learning locally linearized behavior of a non-linear system. We apply the method by learning the parameters of a latent, multivariate Fokker-Planck SDE representing a toggle-switch genetic circuit using biologically realistic parameters, and compare the efficacy of learning relative to the discrete-time Kalman filter as the step-size irregularity and spectral-radius of the dynamics-matrix increases.
Efficient Transfer Learning in Diffusion Models via Adversarial Noise
Abstract
Diffusion Probabilistic Models (DPMs) have demonstrated substantial promise in image generation tasks but heavily rely on the availability of large amounts of training data. Previous works, like GANs, have tackled the limited data problem by transferring pre-trained models learned with sufficient data. However, those methods are hard to be utilized in DPMs since the distinct differences between DPM-based and GAN-based methods, showing in the unique iterative denoising process integral and the need for many timesteps with no-targeted noise in DPMs. In this paper, we propose a novel DPMs-based transfer learning method, TAN, to address the limited data problem. It includes two strategies: similarity-guided training, which boosts transfer with a classifier, and adversarial noise selection which adaptive chooses targeted noise based on the input image. Extensive experiments in the context of few-shot image generation tasks demonstrate that our method is not only efficient but also excels in terms of image quality and diversity when compared to existing GAN-based and DDPM-based methods.
CED: Consistent ensemble distillation for audio tagging
Authors: Heinrich Dinkel, Yongqing Wang, Zhiyong Yan, Junbo Zhang, Yujun Wang
Subjects: Sound (cs.SD); Audio and Speech Processing (eess.AS)
Abstract
Augmentation and knowledge distillation (KD) are well-established techniques employed in the realm of audio classification tasks, aimed at enhancing performance and reducing model sizes on the widely recognized Audioset (AS) benchmark. Although both techniques are effective individually, their combined use, called consistent teaching, hasn't been explored before. This paper proposes CED, a simple training framework that distils student models from large teacher ensembles with consistent teaching. To achieve this, CED efficiently stores logits as well as the augmentation methods on disk, making it scalable to large-scale datasets. Central to CED's efficacy is its label-free nature, meaning that only the stored logits are used for the optimization of a student model only requiring 0.3\% additional disk space for AS. The study trains various transformer-based models, including a 10M parameter model achieving a 49.0 mean average precision (mAP) on AS. Pretrained models and code are available at https://github.com/RicherMans/CED.
EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE
Abstract
Building scalable vision-language models to learn from diverse, multimodal data remains an open challenge. In this paper, we introduce an Efficient Vision-languagE foundation model, namely EVE, which is one unified multimodal Transformer pre-trained solely by one unified pre-training task. Specifically, EVE encodes both vision and language within a shared Transformer network integrated with modality-aware sparse Mixture-of-Experts (MoE) modules, which capture modality-specific information by selectively switching to different experts. To unify pre-training tasks of vision and language, EVE performs masked signal modeling on image-text pairs to reconstruct masked signals, i.e., image pixels and text tokens, given visible signals. This simple yet effective pre-training objective accelerates training by 3.5x compared to the model pre-trained with Image-Text Contrastive and Image-Text Matching losses. Owing to the combination of the unified architecture and pre-training task, EVE is easy to scale up, enabling better downstream performance with fewer resources and faster training speed. Despite its simplicity, EVE achieves state-of-the-art performance on various vision-language downstream tasks, including visual question answering, visual reasoning, and image-text retrieval.
ESTA: An Efficient Spatial-Temporal Range Aggregation Query Processing Algorithm for UAV Networks
Abstract
Unmanned Aerial Vehicle (UAV) networks have been widely used in both military and civilian scenarios. When users are interested in the statistical information of the historical sensory data in a certain region during a certain time period, they will send an aggregation query request with a spatial-temporal constraint to target UAVs which store the qualified data. Then, the target UAVs will return the query results to users. Meanwhile, the query results can be aggregated within the network during transmission to save energy and bandwidth resources, which are typically scarce in UAV networks. However, due to the unique characteristics of UAV networks, it is difficult to perform efficient in-network aggregation of query results without the sacrifice of the user query delay. To the best of our knowledge, there is no research on spatial-temporal range aggregation query in UAV networks. In this paper, we propose an Efficient Spatial-Temporal range Aggregation query processing (ESTA) algorithm for UAV networks. First, a topology change graph is constructed based on the pre-planned trajectory information. Meanwhile, an efficient shortest path algorithm is proposed to obtain the user query delay. Then, on the basis of ensuring the user query delay, ESTA transforms the aggregation processing of query results into recursively solving the set cover problem, thereby constructing a spatial-temporal aggregation tree (STAT), based on which an efficient in-network aggregation routing path for query results can be found. Through extensive simulation, we demonstrate that ESTA can save more than 50% of the energy consumption compared with the baseline algorithm.
RankMixup: Ranking-Based Mixup Training for Network Calibration
Authors: Jongyoun Noh, Hyekang Park, Junghyup Lee, Bumsub Ham
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Network calibration aims to accurately estimate the level of confidences, which is particularly important for employing deep neural networks in real-world systems. Recent approaches leverage mixup to calibrate the network's predictions during training. However, they do not consider the problem that mixtures of labels in mixup may not accurately represent the actual distribution of augmented samples. In this paper, we present RankMixup, a novel mixup-based framework alleviating the problem of the mixture of labels for network calibration. To this end, we propose to use an ordinal ranking relationship between raw and mixup-augmented samples as an alternative supervisory signal to the label mixtures for network calibration. We hypothesize that the network should estimate a higher level of confidence for the raw samples than the augmented ones (Fig.1). To implement this idea, we introduce a mixup-based ranking loss (MRL) that encourages lower confidences for augmented samples compared to raw ones, maintaining the ranking relationship. We also propose to leverage the ranking relationship among multiple mixup-augmented samples to further improve the calibration capability. Augmented samples with larger mixing coefficients are expected to have higher confidences and vice versa (Fig.1). That is, the order of confidences should be aligned with that of mixing coefficients. To this end, we introduce a novel loss, M-NDCG, in order to reduce the number of misaligned pairs of the coefficients and confidences. Extensive experimental results on standard benchmarks for network calibration demonstrate the effectiveness of RankMixup.
StofNet: Super-resolution Time of Flight Network
Authors: Christopher Hahne, Michel Hayoz, Raphael Sznitman
Subjects: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV); Geophysics (physics.geo-ph)
Abstract
Time of Flight (ToF) is a prevalent depth sensing technology in the fields of robotics, medical imaging, and non-destructive testing. Yet, ToF sensing faces challenges from complex ambient conditions making an inverse modelling from the sparse temporal information intractable. This paper highlights the potential of modern super-resolution techniques to learn varying surroundings for a reliable and accurate ToF detection. Unlike existing models, we tailor an architecture for sub-sample precise semi-global signal localization by combining super-resolution with an efficient residual contraction block to balance between fine signal details and large scale contextual information. We consolidate research on ToF by conducting a benchmark comparison against six state-of-the-art methods for which we employ two publicly available datasets. This includes the release of our SToF-Chirp dataset captured by an airborne ultrasound transducer. Results showcase the superior performance of our proposed StofNet in terms of precision, reliability and model complexity. Our code is available at https://github.com/hahnec/stofnet.
Bias-Aware Minimisation: Understanding and Mitigating Estimator Bias in Private SGD
Authors: Moritz Knolle, Robert Dorfman, Alexander Ziller, Daniel Rueckert, Georgios Kaissis
Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
Abstract
Differentially private SGD (DP-SGD) holds the promise of enabling the safe and responsible application of machine learning to sensitive datasets. However, DP-SGD only provides a biased, noisy estimate of a mini-batch gradient. This renders optimisation steps less effective and limits model utility as a result. With this work, we show a connection between per-sample gradient norms and the estimation bias of the private gradient oracle used in DP-SGD. Here, we propose Bias-Aware Minimisation (BAM) that allows for the provable reduction of private gradient estimator bias. We show how to efficiently compute quantities needed for BAM to scale to large neural networks and highlight similarities to closely related methods such as Sharpness-Aware Minimisation. Finally, we provide empirical evidence that BAM not only reduces bias but also substantially improves privacy-utility trade-offs on the CIFAR-10, CIFAR-100, and ImageNet-32 datasets.
MARC: Multipolicy and Risk-aware Contingency Planning for Autonomous Driving
Authors: Tong Li, Lu Zhang, Sikang Liu, Shaojie Shen
Abstract
Generating safe and non-conservative behaviors in dense, dynamic environments remains challenging for automated vehicles due to the stochastic nature of traffic participants' behaviors and their implicit interaction with the ego vehicle. This paper presents a novel planning framework, Multipolicy And Risk-aware Contingency planning (MARC), that systematically addresses these challenges by enhancing the multipolicy-based pipelines from both behavior and motion planning aspects. Specifically, MARC realizes a critical scenario set that reflects multiple possible futures conditioned on each semantic-level ego policy. Then, the generated policy-conditioned scenarios are further formulated into a tree-structured representation with a dynamic branchpoint based on the scene-level divergence. Moreover, to generate diverse driving maneuvers, we introduce risk-aware contingency planning, a bi-level optimization algorithm that simultaneously considers multiple future scenarios and user-defined risk tolerance levels. Owing to the more unified combination of behavior and motion planning layers, our framework achieves efficient decision-making and human-like driving maneuvers. Comprehensive experimental results demonstrate superior performance to other strong baselines in various environments.
Reranking Passages with Coarse-to-Fine Neural Retriever using List-Context Information
Abstract
Passage reranking is a crucial task in many applications, particularly when dealing with large-scale documents. Traditional neural architectures are limited in retrieving the best passage for a question because they usually match the question to each passage separately, seldom considering contextual information in other passages that can provide comparison and reference information. This paper presents a list-context attention mechanism to augment the passage representation by incorporating the list-context information from other candidates. The proposed coarse-to-fine (C2F) neural retriever addresses the out-of-memory limitation of the passage attention mechanism by dividing the list-context modeling process into two sub-processes, allowing for efficient encoding of context information from a large number of candidate answers. This method can be generally used to encode context information from any number of candidate answers in one pass. Different from most multi-stage information retrieval architectures, this model integrates the coarse and fine rankers into the joint optimization process, allowing for feedback between the two layers to update the model simultaneously. Experiments demonstrate the effectiveness of the proposed approach.
IncreLoRA: Incremental Parameter Allocation Method for Parameter-Efficient Fine-tuning
Abstract
With the increasing size of pre-trained language models (PLMs), fine-tuning all the parameters in the model is not efficient, especially when there are a large number of downstream tasks, which incur significant training and storage costs. Many parameter-efficient fine-tuning (PEFT) approaches have been proposed, among which, Low-Rank Adaptation (LoRA) is a representative approach that injects trainable rank decomposition matrices into every target module. Yet LoRA ignores the importance of parameters in different modules. To address this problem, many works have been proposed to prune the parameters of LoRA. However, under limited training conditions, the upper bound of the rank of the pruned parameter matrix is still affected by the preset values. We, therefore, propose IncreLoRA, an incremental parameter allocation method that adaptively adds trainable parameters during training based on the importance scores of each module. This approach is different from the pruning method as it is not limited by the initial number of training parameters, and each parameter matrix has a higher rank upper bound for the same training overhead. We conduct extensive experiments on GLUE to demonstrate the effectiveness of IncreLoRA. The results show that our method owns higher parameter efficiency, especially when under the low-resource settings where our method significantly outperforms the baselines. Our code is publicly available.
A multiobjective continuation method to compute the regularization path of deep neural networks
Authors: Augustina C. Amakor, Konstantin Sontag, Sebastian Peitz
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC); Machine Learning (stat.ML)
Abstract
Sparsity is a highly desired feature in deep neural networks (DNNs) since it ensures numerical efficiency, improves the interpretability of models (due to the smaller number of relevant features), and robustness. In machine learning approaches based on linear models, it is well known that there exists a connecting path between the sparsest solution in terms of the $\ell^1$ norm (i.e., zero weights) and the non-regularized solution, which is called the regularization path. Very recently, there was a first attempt to extend the concept of regularization paths to DNNs by means of treating the empirical loss and sparsity ($\ell^1$ norm) as two conflicting criteria and solving the resulting multiobjective optimization problem. However, due to the non-smoothness of the $\ell^1$ norm and the high number of parameters, this approach is not very efficient from a computational perspective. To overcome this limitation, we present an algorithm that allows for the approximation of the entire Pareto front for the above-mentioned objectives in a very efficient manner. We present numerical examples using both deterministic and stochastic gradients. We furthermore demonstrate that knowledge of the regularization path allows for a well-generalizing network parametrization.
Layer-wise Feedback Propagation
Authors: Leander Weber, Jim Berend, Alexander Binder, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
Abstract
In this paper, we present Layer-wise Feedback Propagation (LFP), a novel training approach for neural-network-like predictors that utilizes explainability, specifically Layer-wise Relevance Propagation(LRP), to assign rewards to individual connections based on their respective contributions to solving a given task. This differs from traditional gradient descent, which updates parameters towards anestimated loss minimum. LFP distributes a reward signal throughout the model without the need for gradient computations. It then strengthens structures that receive positive feedback while reducingthe influence of structures that receive negative feedback. We establish the convergence of LFP theoretically and empirically, and demonstrate its effectiveness in achieving comparable performance to gradient descent on various models and datasets. Notably, LFP overcomes certain limitations associated with gradient-based methods, such as reliance on meaningful derivatives. We further investigate how the different LRP-rules can be extended to LFP, what their effects are on training, as well as potential applications, such as training models with no meaningful derivatives, e.g., step-function activated Spiking Neural Networks (SNNs), or for transfer learning, to efficiently utilize existing knowledge.
InstructionGPT-4: A 200-Instruction Paradigm for Fine-Tuning MiniGPT-4
Authors: Lai Wei, Zihao Jiang, Weiran Huang, Lichao Sun
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Multimodal large language models acquire their instruction-following capabilities through a two-stage training process: pre-training on image-text pairs and fine-tuning on supervised vision-language instruction data. Recent studies have shown that large language models can achieve satisfactory results even with a limited amount of high-quality instruction-following data. In this paper, we introduce InstructionGPT-4, which is fine-tuned on a small dataset comprising only 200 examples, amounting to approximately 6% of the instruction-following data used in the alignment dataset for MiniGPT-4. We first propose several metrics to access the quality of multimodal instruction data. Based on these metrics, we present a simple and effective data selector to automatically identify and filter low-quality vision-language data. By employing this method, InstructionGPT-4 outperforms the original MiniGPT-4 on various evaluations (e.g., visual question answering, GPT-4 preference). Overall, our findings demonstrate that less but high-quality instruction tuning data is efficient to enable multimodal large language models to generate better output.
State Merging with Quantifiers in Symbolic Execution
Authors: David Trabish, Noam Rinetzky, Sharon Shoham, Vaibhav Sharma
Abstract
We address the problem of constraint encoding explosion which hinders the applicability of state merging in symbolic execution. Specifically, our goal is to reduce the number of disjunctions and \emph{if-then-else} expressions introduced during state merging. The main idea is to dynamically partition the symbolic states into merging groups according to a similar uniform structure detected in their path constraints, which allows to efficiently encode the merged path constraint and memory using quantifiers. To address the added complexity of solving quantified constraints, we propose a specialized solving procedure that reduces the solving time in many cases. Our evaluation shows that our approach can lead to significant performance gains.
Using the TypeScript compiler to fix erroneous Node.js snippets
Authors: Brittany Reid, Christoph Treude, Markus Wagner
Abstract
Most online code snippets do not run. This means that developers looking to reuse code from online sources must manually find and fix errors. We present an approach for automatically evaluating and correcting errors in Node.js code snippets: Node Code Correction (NCC). NCC leverages the ability of the TypeScript compiler to generate errors and inform code corrections through the combination of TypeScript's built-in codefixes, our own targeted fixes, and deletion of erroneous lines. Compared to existing approaches using linters, our findings suggest that NCC is capable of detecting a larger number of errors per snippet and more error types, and it is more efficient at fixing snippets. We find that 73.7% of the code snippets in NPM documentation have errors; with the use of NCC's corrections, this number was reduced to 25.1%. Our evaluation confirms that the use of the TypeScript compiler to inform code corrections is a promising strategy to aid in the reuse of code snippets from online sources.
Path-Constrained State Estimation for Rail Vehicles
Authors: Cornelius von Einem, Andrei Cramariuc, Roland Siegwart, Cesar Cadena, Florian Tschopp
Subjects: Robotics (cs.RO); Signal Processing (eess.SP)
Abstract
Globally rising demand for transportation by rail is pushing existing infrastructure to its capacity limits, necessitating the development of accurate, robust, and high-frequency positioning systems to ensure safe and efficient train operation. As individual sensor modalities cannot satisfy the strict requirements of robustness and safety, a combination thereof is required. We propose a path-constrained sensor fusion framework to integrate various modalities while leveraging the unique characteristics of the railway network. To reflect the constrained motion of rail vehicles along their tracks, the state is modeled in 1D along the track geometry. We further leverage the limited action space of a train by employing a novel multi-hypothesis tracking to account for multiple possible trajectories a vehicle can take through the railway network. We demonstrate the reliability and accuracy of our fusion framework on multiple tram datasets recorded in the city of Zurich, utilizing Visual-Inertial Odometry for local motion estimation and a standard GNSS for global localization. We evaluate our results using ground truth localizations recorded with a RTK-GNSS, and compare our method to standard baselines. A Root Mean Square Error of 4.78 m and a track selectivity score of up to 94.9 % have been achieved.
Out of the Cage: How Stochastic Parrots Win in Cyber Security Environments
Authors: Maria Rigaki, Ondřej Lukáš, Carlos A. Catania, Sebastian Garcia
Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Abstract
Large Language Models (LLMs) have gained widespread popularity across diverse domains involving text generation, summarization, and various natural language processing tasks. Despite their inherent limitations, LLM-based designs have shown promising capabilities in planning and navigating open-world scenarios. This paper introduces a novel application of pre-trained LLMs as agents within cybersecurity network environments, focusing on their utility for sequential decision-making processes. We present an approach wherein pre-trained LLMs are leveraged as attacking agents in two reinforcement learning environments. Our proposed agents demonstrate similar or better performance against state-of-the-art agents trained for thousands of episodes in most scenarios and configurations. In addition, the best LLM agents perform similarly to human testers of the environment without any additional training process. This design highlights the potential of LLMs to efficiently address complex decision-making tasks within cybersecurity. Furthermore, we introduce a new network security environment named NetSecGame. The environment is designed to eventually support complex multi-agent scenarios within the network security domain. The proposed environment mimics real network attacks and is designed to be highly modular and adaptable for various scenarios.
Multi-UAV Deployment in Obstacle-Cluttered Environments with LOS Connectivity
Abstract
A reliable communication network is essential for multiple UAVs operating within obstacle-cluttered environments, where limited communication due to obstructions often occurs. A common solution is to deploy intermediate UAVs to relay information via a multi-hop network, which introduces two challenges: (i) how to design the structure of multi-hop networks; and (ii) how to maintain connectivity during collaborative motion. To this end, this work first proposes an efficient constrained search method based on the minimum-edge RRT$^\star$ algorithm, to find a spanning-tree topology that requires a less number of UAVs for the deployment task. To achieve this deployment, a distributed model predictive control strategy is proposed for the online motion coordination. It explicitly incorporates not only the inter-UAV and UAV-obstacle distance constraints, but also the line-of-sight (LOS) connectivity constraint. These constraints are well-known to be nonlinear and often tackled by various approximations. In contrast, this work provides a theoretical guarantee that all agent trajectories are ensured to be collision-free with a team-wise LOS connectivity at all time. Numerous simulations are performed in 3D valley-like environments, while hardware experiments validate its dynamic adaptation when the deployment position changes online.
DarkDiff: Explainable web page similarity of TOR onion sites
Authors: Pieter Hartel, Eljo Haspels, Mark van Staalduinen, Octavio Texeira
Abstract
In large-scale data analysis, near-duplicates are often a problem. For example, with two near-duplicate phishing emails, a difference in the salutation (Mr versus Ms) is not essential, but whether it is bank A or B is important. The state-of-the-art in near-duplicate detection is a black box approach (MinHash), so one only knows that emails are near-duplicates, but not why. We present DarkDiff, which can efficiently detect near-duplicates while providing the reason why there is a near-duplicate. We have developed DarkDiff to detect near-duplicates of homepages on the Darkweb. DarkDiff works well on those pages because they resemble the clear web of the past.
Modeling excitable cells with the EMI equations: spectral analysis and iterative solution strategy
Authors: Pietro Benedusi, Paola Ferrari, Marie Rognes, Stefano Serra-Capizzano
Abstract
In this work, we are interested in solving large linear systems stemming from the Extra-Membrane-Intra (EMI) model, which is employed for simulating excitable tissues at a cellular scale. After setting the related systems of partial differential equations (PDEs) equipped with proper boundary conditions, we provide numerical approximation schemes for the EMI PDEs and focus on the resulting large linear systems. We first give a relatively complete spectral analysis using tools from the theory of Generalized Locally Toeplitz matrix sequences. The obtained spectral information is used for designing appropriate (preconditioned) Krylov solvers. We show, through numerical experiments, that the presented solution strategy is robust w.r.t. problem and discretization parameters, efficient and scalable.
Unsupervised anomalies detection in IIoT edge devices networks using federated learning
Abstract
In a connection of many IoT devices that each collect data, normally training a machine learning model would involve transmitting the data to a central server which requires strict privacy rules. However, some owners are reluctant of availing their data out of the company due to data security concerns. Federated learning(FL) as a distributed machine learning approach performs training of a machine learning model on the device that gathered the data itself. In this scenario, data is not share over the network for training purpose. Fedavg as one of FL algorithms permits a model to be copied to participating devices during a training session. The devices could be chosen at random, and a device can be aborted. The resulting models are sent to the coordinating server and then average models from the devices that finished training. The process is repeated until a desired model accuracy is achieved. By doing this, FL approach solves the privacy problem for IoT/ IIoT devices that held sensitive data for the owners. In this paper, we leverage the benefits of FL and implemented Fedavg algorithm on a recent dataset that represent the modern IoT/ IIoT device networks. The results were almost the same as the centralized machine learning approach. We also evaluated some shortcomings of Fedavg such as unfairness that happens during the training when struggling devices do not participate for every stage of training. This inefficient training of local or global model could lead in a high number of false alarms in intrusion detection systems for IoT/IIoT gadgets developed using Fedavg. Hence, after evaluating the FedAv deep auto encoder with centralized deep auto encoder ML, we further proposed and designed a Fair Fedavg algorithm that will be evaluated in the future work.
Curriculum Learning with Adam: The Devil Is in the Wrong Details
Authors: Lucas Weber, Jaap Jumelet, Paul Michel, Elia Bruni, Dieuwke Hupkes
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Abstract
Curriculum learning (CL) posits that machine learning models -- similar to humans -- may learn more efficiently from data that match their current learning progress. However, CL methods are still poorly understood and, in particular for natural language processing (NLP), have achieved only limited success. In this paper, we explore why. Starting from an attempt to replicate and extend a number of recent curriculum methods, we find that their results are surprisingly brittle when applied to NLP. A deep dive into the (in)effectiveness of the curricula in some scenarios shows us why: when curricula are employed in combination with the popular Adam optimisation algorithm, they oftentimes learn to adapt to suboptimally chosen optimisation parameters for this algorithm. We present a number of different case studies with different common hand-crafted and automated CL approaches to illustrate this phenomenon, and we find that none of them outperforms optimisation with only Adam with well-chosen hyperparameters. As such, our results contribute to understanding why CL methods work, but at the same time urge caution when claiming positive results.
A Heuristic Informative-Path-Planning Algorithm for Autonomous Mapping of Unknown Areas
Authors: Mobolaji O. Orisatoki, Mahdi Amouzadi, Arash M. Dizqah
Abstract
Informative path planning algorithms are of paramount importance in applications like disaster management to efficiently gather information through a priori unknown environments. This is, however, a complex problem that involves finding a globally optimal path that gathers the maximum amount of information (e.g., the largest map with a minimum travelling distance) while using partial and uncertain local measurements. This paper addresses this problem by proposing a novel heuristic algorithm that continuously estimates the potential mapping gain for different sub-areas across the partially created map, and then uses these estimations to locally navigate the robot. Furthermore, this paper presents a novel algorithm to calculate a benchmark solution, where the map is a priori known to the planar, to evaluate the efficacy of the developed heuristic algorithm over different test scenarios. The findings indicate that the efficiency of the proposed algorithm, measured in terms of the mapped area per unit of travelling distance, ranges from 70% to 80% of the benchmark solution in various test scenarios. In essence, the algorithm demonstrates the capability to generate paths that come close to the globally optimal path provided by the benchmark solution.
How to Protect Copyright Data in Optimization of Large Language Models?
Authors: Timothy Chu, Zhao Song, Chiwun Yang
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Abstract
Large language models (LLMs) and generative AI have played a transformative role in computer research and applications. Controversy has arisen as to whether these models output copyrighted data, which can occur if the data the models are trained on is copyrighted. LLMs are built on the transformer neural network architecture, which in turn relies on a mathematical computation called Attention that uses the softmax function. In this paper, we show that large language model training and optimization can be seen as a softmax regression problem. We then establish a method of efficiently performing softmax regression, in a way that prevents the regression function from generating copyright data. This establishes a theoretical method of training large language models in a way that avoids generating copyright data.
Language Reward Modulation for Pretraining Reinforcement Learning
Authors: Ademi Adeniji, Amber Xie, Carmelo Sferrazza, Younggyo Seo, Stephen James, Pieter Abbeel
Abstract
Using learned reward functions (LRFs) as a means to solve sparse-reward reinforcement learning (RL) tasks has yielded some steady progress in task-complexity through the years. In this work, we question whether today's LRFs are best-suited as a direct replacement for task rewards. Instead, we propose leveraging the capabilities of LRFs as a pretraining signal for RL. Concretely, we propose $\textbf{LA}$nguage Reward $\textbf{M}$odulated $\textbf{P}$retraining (LAMP) which leverages the zero-shot capabilities of Vision-Language Models (VLMs) as a $\textit{pretraining}$ utility for RL as opposed to a downstream task reward. LAMP uses a frozen, pretrained VLM to scalably generate noisy, albeit shaped exploration rewards by computing the contrastive alignment between a highly diverse collection of language instructions and the image observations of an agent in its pretraining environment. LAMP optimizes these rewards in conjunction with standard novelty-seeking exploration rewards with reinforcement learning to acquire a language-conditioned, pretrained policy. Our VLM pretraining approach, which is a departure from previous attempts to use LRFs, can warmstart sample-efficient learning on robot manipulation tasks in RLBench.
Keyword: faster
Utilizing Admissible Bounds for Heuristic Learning
Abstract
While learning a heuristic function for forward search algorithms with modern machine learning techniques has been gaining interest in recent years, there has been little theoretical understanding of \emph{what} they should learn, \emph{how} to train them, and \emph{why} we do so. This lack of understanding leads to various literature performing an ad-hoc selection of datasets (suboptimal vs optimal costs or admissible vs inadmissible heuristics) and optimization metrics (e.g., squared vs absolute errors). Moreover, due to the lack of admissibility of the resulting trained heuristics, little focus has been put on the role of admissibility \emph{during} learning. This paper articulates the role of admissible heuristics in supervised heuristic learning using them as parameters of Truncated Gaussian distributions, which tightens the hypothesis space compared to ordinary Gaussian distributions. We argue that this mathematical model faithfully follows the principle of maximum entropy and empirically show that, as a result, it yields more accurate heuristics and converges faster during training.
EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE
Abstract
Building scalable vision-language models to learn from diverse, multimodal data remains an open challenge. In this paper, we introduce an Efficient Vision-languagE foundation model, namely EVE, which is one unified multimodal Transformer pre-trained solely by one unified pre-training task. Specifically, EVE encodes both vision and language within a shared Transformer network integrated with modality-aware sparse Mixture-of-Experts (MoE) modules, which capture modality-specific information by selectively switching to different experts. To unify pre-training tasks of vision and language, EVE performs masked signal modeling on image-text pairs to reconstruct masked signals, i.e., image pixels and text tokens, given visible signals. This simple yet effective pre-training objective accelerates training by 3.5x compared to the model pre-trained with Image-Text Contrastive and Image-Text Matching losses. Owing to the combination of the unified architecture and pre-training task, EVE is easy to scale up, enabling better downstream performance with fewer resources and faster training speed. Despite its simplicity, EVE achieves state-of-the-art performance on various vision-language downstream tasks, including visual question answering, visual reasoning, and image-text retrieval.
Towards Privacy-Supporting Fall Detection via Deep Unsupervised RGB2Depth Adaptation
Abstract
Fall detection is a vital task in health monitoring, as it allows the system to trigger an alert and therefore enabling faster interventions when a person experiences a fall. Although most previous approaches rely on standard RGB video data, such detailed appearance-aware monitoring poses significant privacy concerns. Depth sensors, on the other hand, are better at preserving privacy as they merely capture the distance of objects from the sensor or camera, omitting color and texture information. In this paper, we introduce a privacy-supporting solution that makes the RGB-trained model applicable in depth domain and utilizes depth data at test time for fall detection. To achieve cross-modal fall detection, we present an unsupervised RGB to Depth (RGB2Depth) cross-modal domain adaptation approach that leverages labelled RGB data and unlabelled depth data during training. Our proposed pipeline incorporates an intermediate domain module for feature bridging, modality adversarial loss for modality discrimination, classification loss for pseudo-labeled depth data and labeled source data, triplet loss that considers both source and target domains, and a novel adaptive loss weight adjustment method for improved coordination among various losses. Our approach achieves state-of-the-art results in the unsupervised RGB2Depth domain adaptation task for fall detection. Code is available at https://github.com/1015206533/privacy_supporting_fall_detection.
Keyword: mobile
Towards an On-device Agent for Text Rewriting
Authors: Yun Zhu, Yinxiao Liu, Felix Stahlberg, Shankar Kumar, Yu-hui Chen, Liangchen Luo, Lei Shu, Renjie Liu, Jindong Chen, Lei Meng
Abstract
Large Language Models (LLMs) have demonstrated impressive capabilities for text rewriting. Nonetheless, the large sizes of these models make them impractical for on-device inference, which would otherwise allow for enhanced privacy and economical inference. Creating a smaller yet potent language model for text rewriting presents a formidable challenge because it requires balancing the need for a small size with the need to retain the emergent capabilities of the LLM, that requires costly data collection. To address the above challenge, we introduce a new instruction tuning approach for building a mobile-centric text rewriting model. Our strategies enable the generation of high quality training data without any human labeling. In addition, we propose a heuristic reinforcement learning framework which substantially enhances performance without requiring preference data. To further bridge the performance gap with the larger server-side model, we propose an effective approach that combines the mobile rewrite agent with the server model using a cascade. To tailor the text rewriting tasks to mobile scenarios, we introduce MessageRewriteEval, a benchmark that focuses on text rewriting for messages through natural language instructions. Through empirical experiments, we demonstrate that our on-device model surpasses the current state-of-the-art LLMs in text rewriting while maintaining a significantly reduced model size. Notably, we show that our proposed cascading approach improves model performance.
PatchBackdoor: Backdoor Attack against Deep Neural Networks without Model Modification
Authors: Yizhen Yuan (1), Rui Kong (3), Shenghao Xie (4), Yuanchun Li (1 and 2), Yunxin Liu (1 and 2) ((1) Institute for AI Industry Research (AIR), Tsinghua University, Beijing, China, (2) Shanghai AI Laboratory, Shanghai, China, (3) Shanghai Jiao Tong University, Shanghai, China, (4) Wuhan University, Wuhan, China)
Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Backdoor attack is a major threat to deep learning systems in safety-critical scenarios, which aims to trigger misbehavior of neural network models under attacker-controlled conditions. However, most backdoor attacks have to modify the neural network models through training with poisoned data and/or direct model editing, which leads to a common but false belief that backdoor attack can be easily avoided by properly protecting the model. In this paper, we show that backdoor attacks can be achieved without any model modification. Instead of injecting backdoor logic into the training data or the model, we propose to place a carefully-designed patch (namely backdoor patch) in front of the camera, which is fed into the model together with the input images. The patch can be trained to behave normally at most of the time, while producing wrong prediction when the input image contains an attacker-controlled trigger object. Our main techniques include an effective training method to generate the backdoor patch and a digital-physical transformation modeling method to enhance the feasibility of the patch in real deployments. Extensive experiments show that PatchBackdoor can be applied to common deep learning models (VGG, MobileNet, ResNet) with an attack success rate of 93% to 99% on classification tasks. Moreover, we implement PatchBackdoor in real-world scenarios and show that the attack is still threatening.
A deep reinforcement learning approach for real-time demand-responsive railway rescheduling to mitigate station overcrowding using mobile data
Authors: Enze Liu, Zhiyuan Lin, Judith Y.T. Wang, Hong Chen
Subjects: Systems and Control (eess.SY); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Abstract
Real-time railway rescheduling is a timely and flexible technique to automatically alter the operation schedule in response to time-varying conditions. Current research lacks data-driven approaches that capture real-time passenger mobility during railway disruptions, relying mostly on OD-based data and model-based methods for estimating demands of trains. Meanwhile, the schedule-updating principles for a long-term disruption overlook the uneven distribution of demand over time. To fill this gap, this paper proposes a demand-responsive approach by inferring real-world passenger mobility from mobile data (MD) to facilitate real-time rescheduling. Unlike network-level approaches, this paper focuses on a heavy-demand station upstream of the disrupted area. The objective is to reschedule all trains on multiple routes passing through this target station, which have been affected by a severe emergency event such as a natural disaster. Particular attention should be given to avoiding the accumulation of overcrowded passengers at this station, to prevent additional accidents arising from overcrowding. This research addresses the challenges associated with this scenario, including the dynamics of arriving and leaving of passengers, station overcrowding, rolling stock shortage, open-ended disruption duration, integrated rescheduling on multiple routes, and delays due to detours. A deep reinforcement learning (DRL) framework is proposed to determine the optimal rescheduled timetable, route stops, and rolling stock allocation, while considering real-time demand satisfaction, station overcrowding, train capacity utilization, and headway safety.
Abstract
Mobile robotic agents often suffer from localization uncertainty which grows with time and with the agents' movement. This can hinder their ability to accomplish their task. In some settings, it may be possible to perform assistive actions that reduce uncertainty about a robot's location. For example, in a collaborative multi-robot system, a wheeled robot can request assistance from a drone that can fly to its estimated location and reveal its exact location on the map or accompany it to its intended location. Since assistance may be costly and limited, and may be requested by different members of a team, there is a need for principled ways to support the decision of which assistance to provide to an agent and when, as well as to decide which agent to help within a team. For this purpose, we propose Value of Assistance (VOA) to represent the expected cost reduction that assistance will yield at a given point of execution. We offer ways to compute VOA based on estimations of the robot's future uncertainty, modeled as a Gaussian process. We specify conditions under which our VOA measures are valid and empirically demonstrate the ability of our measures to predict the agent's average cost reduction when receiving assistance in both simulated and real-world robotic settings.
Keyword: pruning
Adaptive Encoding Strategies for Erasing-Based Lossless Floating-Point Compression
Abstract
Lossless floating-point time series compression is crucial for a wide range of critical scenarios. Nevertheless, it is a big challenge to compress time series losslessly due to the complex underlying layouts of floating-point values. The state-of-the-art erasing-based compression algorithm Elf demonstrates a rather impressive performance. We give an in-depth exploration of the encoding strategies of Elf, and find that there is still much room for improvement. In this paper, we propose Elf, which employs a set of optimizations for leading zeros, center bits and sharing condition. Specifically, we develop a dynamic programming algorithm with a set of pruning strategies to compute the adaptive approximation rules efficiently. We theoretically prove that the adaptive approximation rules are globally optimal. We further extend Elf to Streaming Elf, i.e., SElf, which achieves almost the same compression ratio as Elf, while enjoying even higher efficiency in streaming scenarios. We compare Elf and SElf with 8 competitors using 22 datasets. The results demonstrate that SElf achieves 9.2% relative compression ratio improvement over the best streaming competitor while maintaining similar efficiency, and that Elf* ranks among the most competitive batch compressors. All source codes are publicly released.
IncreLoRA: Incremental Parameter Allocation Method for Parameter-Efficient Fine-tuning
Abstract
With the increasing size of pre-trained language models (PLMs), fine-tuning all the parameters in the model is not efficient, especially when there are a large number of downstream tasks, which incur significant training and storage costs. Many parameter-efficient fine-tuning (PEFT) approaches have been proposed, among which, Low-Rank Adaptation (LoRA) is a representative approach that injects trainable rank decomposition matrices into every target module. Yet LoRA ignores the importance of parameters in different modules. To address this problem, many works have been proposed to prune the parameters of LoRA. However, under limited training conditions, the upper bound of the rank of the pruned parameter matrix is still affected by the preset values. We, therefore, propose IncreLoRA, an incremental parameter allocation method that adaptively adds trainable parameters during training based on the importance scores of each module. This approach is different from the pruning method as it is not limited by the initial number of training parameters, and each parameter matrix has a higher rank upper bound for the same training overhead. We conduct extensive experiments on GLUE to demonstrate the effectiveness of IncreLoRA. The results show that our method owns higher parameter efficiency, especially when under the low-resource settings where our method significantly outperforms the baselines. Our code is publicly available.
Keyword: diffusion
Revolutionizing TCAD Simulations with Universal Device Encoding and Graph Attention Networks
Abstract
An innovative methodology that leverages artificial intelligence (AI) and graph representation for semiconductor device encoding in TCAD device simulation is proposed. A graph-based universal encoding scheme is presented that not only considers material-level and device-level embeddings, but also introduces a novel spatial relationship embedding inspired by interpolation operations typically used in finite element meshing. Universal physical laws from device simulations are leveraged for comprehensive data-driven modeling, which encompasses surrogate Poisson emulation and current-voltage (IV) prediction based on drift-diffusion model. Both are achieved using a novel graph attention network, referred to as RelGAT. Comprehensive technical details based on the device simulator Sentaurus TCAD are presented, empowering researchers to adopt the proposed AI-driven Electronic Design Automation (EDA) solution at the device level.
The macroscopic finite-difference scheme and modified equations of the general propagation multiple-relaxation-time lattice Boltzmann model
Authors: Ying Chen, Xi Liu, Zhenhua Chai, Baochang Shi
Subjects: Numerical Analysis (math.NA); Analysis of PDEs (math.AP)
Abstract
In this paper, we first present the general propagation multiple-relaxation-time lattice Boltzmann (GPMRT-LB) model and obtain the corresponding macroscopic finite-difference (GPMFD) scheme on conservative moments. Then based on the Maxwell iteration method, we conduct the analysis on the truncation errors and modified equations (MEs) of the GPMRT-LB model and GPMFD scheme at both diffusive and acoustic scalings. For the nonlinear anisotropic convection-diffusion equation (NACDE) and Navier-Stokes equations (NSEs), we also derive the first- and second-order MEs of the GPMRT-LB model and GPMFD scheme. In particular, for the one-dimensional convection-diffusion equation (CDE) with the constant velocity and diffusion coefficient, we can develop a fourth-order GPMRT-LB (F-GPMRT-LB) model and the corresponding fourth-order GPMFD (F-GPMFD) scheme at the diffusive scaling. Finally, two benchmark problems, Gauss hill problem and Poiseuille flow in two-dimensional space, are used to test the GPMRT-LB model and GPMFD scheme, and it is found that the numerical results are not only in good agreement with corresponding analytical solutions, but also have a second-order convergence rate in space. Additionally, a numerical study on one-dimensional CDE also demonstrates that the F-GPMRT-LB model and F-GPMFD scheme can achieve a fourth-order accuracy in space, which is consistent with our theoretical analysis.
Shape-conditioned 3D Molecule Generation via Equivariant Diffusion Models
Authors: Ziqi Chen, Bo Peng, Srinivasan Parthasarathy, Xia Ning
Abstract
Ligand-based drug design aims to identify novel drug candidates of similar shapes with known active molecules. In this paper, we formulated an in silico shape-conditioned molecule generation problem to generate 3D molecule structures conditioned on the shape of a given molecule. To address this problem, we developed a translation- and rotation-equivariant shape-guided generative model ShapeMol. ShapeMol consists of an equivariant shape encoder that maps molecular surface shapes into latent embeddings, and an equivariant diffusion model that generates 3D molecules based on these embeddings. Experimental results show that ShapeMol can generate novel, diverse, drug-like molecules that retain 3D molecular shapes similar to the given shape condition. These results demonstrate the potential of ShapeMol in designing drug candidates of desired 3D shapes binding to protein target pockets.
Boosting Diffusion Models with an Adaptive Momentum Sampler
Abstract
Diffusion probabilistic models (DPMs) have been shown to generate high-quality images without the need for delicate adversarial training. However, the current sampling process in DPMs is prone to violent shaking. In this paper, we present a novel reverse sampler for DPMs inspired by the widely-used Adam optimizer. Our proposed sampler can be readily applied to a pre-trained diffusion model, utilizing momentum mechanisms and adaptive updating to smooth the reverse sampling process and ensure stable generation, resulting in outputs of enhanced quality. By implicitly reusing update directions from early steps, our proposed sampler achieves a better balance between high-level semantics and low-level details. Additionally, this sampler is flexible and can be easily integrated into pre-trained DPMs regardless of the sampler used during training. Our experimental results on multiple benchmarks demonstrate that our proposed reverse sampler yields remarkable improvements over different baselines. We will make the source code available.
LongDanceDiff: Long-term Dance Generation with Conditional Diffusion Model
Authors: Siqi Yang, Zejun Yang, Zhisheng Wang
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
Dancing with music is always an essential human art form to express emotion. Due to the high temporal-spacial complexity, long-term 3D realist dance generation synchronized with music is challenging. Existing methods suffer from the freezing problem when generating long-term dances due to error accumulation and training-inference discrepancy. To address this, we design a conditional diffusion model, LongDanceDiff, for this sequence-to-sequence long-term dance generation, addressing the challenges of temporal coherency and spatial constraint. LongDanceDiff contains a transformer-based diffusion model, where the input is a concatenation of music, past motions, and noised future motions. This partial noising strategy leverages the full-attention mechanism and learns the dependencies among music and past motions. To enhance the diversity of generated dance motions and mitigate the freezing problem, we introduce a mutual information minimization objective that regularizes the dependency between past and future motions. We also address common visual quality issues in dance generation, such as foot sliding and unsmooth motion, by incorporating spatial constraints through a Global-Trajectory Modulation (GTM) layer and motion perceptual losses, thereby improving the smoothness and naturalness of motion generation. Extensive experiments demonstrate a significant improvement in our approach over the existing state-of-the-art methods. We plan to release our codes and models soon.
Efficient Transfer Learning in Diffusion Models via Adversarial Noise
Abstract
Diffusion Probabilistic Models (DPMs) have demonstrated substantial promise in image generation tasks but heavily rely on the availability of large amounts of training data. Previous works, like GANs, have tackled the limited data problem by transferring pre-trained models learned with sufficient data. However, those methods are hard to be utilized in DPMs since the distinct differences between DPM-based and GAN-based methods, showing in the unique iterative denoising process integral and the need for many timesteps with no-targeted noise in DPMs. In this paper, we propose a novel DPMs-based transfer learning method, TAN, to address the limited data problem. It includes two strategies: similarity-guided training, which boosts transfer with a classifier, and adversarial noise selection which adaptive chooses targeted noise based on the input image. Extensive experiments in the context of few-shot image generation tasks demonstrate that our method is not only efficient but also excels in terms of image quality and diversity when compared to existing GAN-based and DDPM-based methods.
High-quality Image Dehazing with Diffusion Model
Authors: Hu Yu, Jie Huang, Kaiwen Zheng, Man Zhou, Feng Zhao
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
Image dehazing is quite challenging in dense-haze scenarios, where quite less original information remains in the hazy image. Though previous methods have made marvelous progress, they still suffer from information loss in content and color in dense-haze scenarios. The recently emerged Denoising Diffusion Probabilistic Model (DDPM) exhibits strong generation ability, showing potential for solving this problem. However, DDPM fails to consider the physics property of dehazing task, limiting its information completion capacity. In this work, we propose DehazeDDPM: A DDPM-based and physics-aware image dehazing framework that applies to complex hazy scenarios. Specifically, DehazeDDPM works in two stages. The former stage physically models the dehazing task with the Atmospheric Scattering Model (ASM), pulling the distribution closer to the clear data and endowing DehazeDDPM with fog-aware ability. The latter stage exploits the strong generation ability of DDPM to compensate for the haze-induced huge information loss, by working in conjunction with the physical modelling. Extensive experiments demonstrate that our method attains state-of-the-art performance on both synthetic and real-world hazy datasets.
Manipulating Embeddings of Stable Diffusion Prompts
Authors: Niklas Deckers, Julia Peters, Martin Potthast
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Abstract
Generative text-to-image models such as Stable Diffusion allow users to generate images based on a textual description, the prompt. Changing the prompt is still the primary means for the user to change a generated image as desired. However, changing the image by reformulating the prompt remains a difficult process of trial and error, which has led to the emergence of prompt engineering as a new field of research. We propose and analyze methods to change the embedding of a prompt directly instead of the prompt text. It allows for more fine-grained and targeted control that takes into account user intentions. Our approach treats the generative text-to-image model as a continuous function and passes gradients between the image space and the prompt embedding space. By addressing different user interaction problems, we can apply this idea in three scenarios: (1) Optimization of a metric defined in image space that could measure, for example, image style. (2) Assistance of users in creative tasks by enabling them to navigate the image space along a selection of directions of "near" prompt embeddings. (3) Changing the embedding of the prompt to include information that the user has seen in a particular seed but finds difficult to describe in the prompt. Our experiments demonstrate the feasibility of the described methods.
Space-time hybridizable discontinuous Galerkin method for advection-diffusion on deforming domains: The advection-dominated regime
Abstract
We analyze a space-time hybridizable discontinuous Galerkin method to solve the time-dependent advection-diffusion equation on deforming domains. We prove stability of the discretization in the advection-dominated regime by using weighted test functions and derive a priori space-time error estimates. A numerical example illustrates the theoretical results.
Diffusion Language Models Can Perform Many Tasks with Scaling and Instruction-Finetuning
Abstract
The recent surge of generative AI has been fueled by the generative power of diffusion probabilistic models and the scalable capabilities of large language models. Despite their potential, it remains elusive whether diffusion language models can solve general language tasks comparable to their autoregressive counterparts. This paper demonstrates that scaling diffusion models w.r.t. data, sizes, and tasks can effectively make them strong language learners. We build competent diffusion language models at scale by first acquiring knowledge from massive data via masked language modeling pretraining thanks to their intrinsic connections. We then reprogram pretrained masked language models into diffusion language models via diffusive adaptation, wherein task-specific finetuning and instruction finetuning are explored to unlock their versatility in solving general language tasks. Experiments show that scaling diffusion language models consistently improves performance across downstream language tasks. We further discover that instruction finetuning can elicit zero-shot and few-shot in-context learning abilities that help tackle many unseen tasks by following natural language instructions, and show promise in advanced and challenging abilities such as reasoning
On-Manifold Projected Gradient Descent
Authors: Aaron Mahler, Tyrus Berry, Tom Stephens, Harbir Antil, Michael Merritt, Jeanie Schreiber, Ioannis Kevrekidis
Abstract
This work provides a computable, direct, and mathematically rigorous approximation to the differential geometry of class manifolds for high-dimensional data, along with nonlinear projections from input space onto these class manifolds. The tools are applied to the setting of neural network image classifiers, where we generate novel, on-manifold data samples, and implement a projected gradient descent algorithm for on-manifold adversarial training. The susceptibility of neural networks (NNs) to adversarial attack highlights the brittle nature of NN decision boundaries in input space. Introducing adversarial examples during training has been shown to reduce the susceptibility of NNs to adversarial attack; however, it has also been shown to reduce the accuracy of the classifier if the examples are not valid examples for that class. Realistic "on-manifold" examples have been previously generated from class manifolds in the latent of an autoencoder. Our work explores these phenomena in a geometric and computational setting that is much closer to the raw, high-dimensional input space than can be provided by VAE or other black box dimensionality reductions. We employ conformally invariant diffusion maps (CIDM) to approximate class manifolds in diffusion coordinates, and develop the Nystr\"{o}m projection to project novel points onto class manifolds in this setting. On top of the manifold approximation, we leverage the spectral exterior calculus (SEC) to determine geometric quantities such as tangent vectors of the manifold. We use these tools to obtain adversarial examples that reside on a class manifold, yet fool a classifier. These misclassifications then become explainable in terms of human-understandable manipulations within the data, by expressing the on-manifold adversary in the semantic basis on the manifold.
Keyword: adaptive
Software-based signal compression algorithm for ROM-stored electrical cables
Authors: Tshimankinda Jerome Ngoy, Mike Nkongolo
Subjects: Information Theory (cs.IT); Hardware Architecture (cs.AR); Signal Processing (eess.SP)
Abstract
This project introduces a groundbreaking approach to address the challenge of periodic signal compression. By proposing a novel adaptive coding method, coupled with hardware-assisted data compression, we have developed a new architecture model tailored for efficient data compression. The selected compression scheme has demonstrated remarkable results, showcasing reduced memory communication volume and power consumption in the cache memory path of benchmark systems. With a reduction range of 4.2% to 35.2%, this innovation paves the way for affordable smart sensing, monitoring, diagnostics, and protection in emerging low-cost device types. Consequently, this cutting-edge technology enhances electrical signal compression and contributes to grid improvement. Additionally, we explore the novel application of harnessing wasted thermal energy in the Read-Only Memory (ROM) using thermoelectricity (TE). This approach captures the excess thermal energy, converting it into electrical energy through optimized supercapacitor charging, resulting in efficient energy utilization. This innovation intersects the fields of embedded systems, data compression, energy efficiency, and smart grid technology.
Reinforcement Learning -based Adaptation and Scheduling Methods for Multi-source DASH
Authors: Nghia T. Nguyen, Long Luu, Phuong L. Vo, Thi Thanh Sang Nguyen, Cuong T. Do, Ngoc-thanh Nguyen
Subjects: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Abstract
Dynamic adaptive streaming over HTTP (DASH) has been widely used in video streaming recently. In DASH, the client downloads video chunks in order from a server. The rate adaptation function at the video client enhances the user's quality-of-experience (QoE) by choosing a suitable quality level for each video chunk to download based on the network condition. Today networks such as content delivery networks, edge caching networks, content-centric networks,... usually replicate video contents on multiple cache nodes. We study video streaming from multiple sources in this work. In multi-source streaming, video chunks may arrive out of order due to different conditions of the network paths. Hence, to guarantee a high QoE, the video client needs not only rate adaptation but also chunk scheduling. Reinforcement learning (RL) has emerged as the state-of-the-art control method in various fields in recent years. This paper proposes two algorithms for streaming from multiple sources: RL-based adaptation with greedy scheduling (RLAGS) and RL-based adaptation and scheduling (RLAS). We also build a simulation environment for training and evaluating. The efficiency of the proposed algorithms is proved via extensive simulations with real-trace data.
A General Framework for Progressive Data Compression and Retrieval
Abstract
In scientific simulations, observations, and experiments, the cost of transferring data to and from disk and across networks has become a significant bottleneck that particularly impacts subsequent data analysis and visualization. To address this challenge, compression techniques have been widely adopted. However, traditional lossy compression approaches often require setting error tolerances conservatively to respect the numerical sensitivities of a wide variety of post hoc data analyses, some of which may not even be known a priori. Progressive data compression and retrieval has emerged as a solution, allowing for the adaptive handling of compressed data according to the needs of a given post-processing task. However, few analysis algorithms natively support progressive data processing, and adapting compression techniques, file formats, client/server frameworks, and APIs to support progressivity can be challenging. This work presents a general framework that supports progressive-precision data queries independently of the underlying data compressor or number representation. Our approach is based on a multiple-component representation that successively, with each new component, reduces the error between the original and compressed field, allowing each field in the progressive sequence to be expressed as a partial sum of components. We have implemented our approach on top of four popular scientific data compressors and have evaluated its behavior on several real-world data sets from the SDRBench collection. Numerical results indicate that our framework is effective in terms of accuracy compared to each of the standalone compressors it builds upon. In addition, (de)compression time is proportional to the number and granularity of components. Finally, our framework allows for fully lossless compression using lossy compressors when a sufficient number of components are employed.
Bayesian Learning for Dynamic Target Localization with Human-provided Spatial Information
Abstract
This paper considers a human-autonomy collaborative sensor data fusion for dynamic target localization in a Bayesian framework. To compensate for the shortcomings of an autonomous tracking system, we propose to collect spatial sensing information from human operators who visually monitor the target and can provide target localization information in the form of free sketches encircling the area where the target is located. Our focus in this paper is to construct an adaptive probabilistic model for human-provided inputs where the adaption terms capture the level of reliability of the human inputs. The next contribution of this paper is a novel joint Bayesian learning method to fuse human and autonomous sensor inputs in a manner that the dynamic changes in human detection reliability are also captured and accounted for. A unique aspect of this Bayesian modeling framework is its analytical closed-form update equations, endowing our method with significant computational efficiency. Simulations demonstrate our results.
SEA: Shareable and Explainable Attribution for Query-based Black-box Attacks
Authors: Yue Gao, Ilia Shumailov, Kassem Fawaz
Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
Abstract
Machine Learning (ML) systems are vulnerable to adversarial examples, particularly those from query-based black-box attacks. Despite various efforts to detect and prevent such attacks, there is a need for a more comprehensive approach to logging, analyzing, and sharing evidence of attacks. While classic security benefits from well-established forensics and intelligence sharing, Machine Learning is yet to find a way to profile its attackers and share information about them. In response, this paper introduces SEA, a novel ML security system to characterize black-box attacks on ML systems for forensic purposes and to facilitate human-explainable intelligence sharing. SEA leverages the Hidden Markov Models framework to attribute the observed query sequence to known attacks. It thus understands the attack's progression rather than just focusing on the final adversarial examples. Our evaluations reveal that SEA is effective at attack attribution, even on their second occurrence, and is robust to adaptive strategies designed to evade forensics analysis. Interestingly, SEA's explanations of the attack behavior allow us even to fingerprint specific minor implementation bugs in attack libraries. For example, we discover that the SignOPT and Square attacks implementation in ART v1.14 sends over 50% specific zero difference queries. We thoroughly evaluate SEA on a variety of settings and demonstrate that it can recognize the same attack's second occurrence with 90+% Top-1 and 95+% Top-3 accuracy.
Multi-object Detection, Tracking and Prediction in Rugged Dynamic Environments
Abstract
Multi-object tracking (MOT) has important applications in monitoring, logistics, and other fields. This paper develops a real-time multi-object tracking and prediction system in rugged environments. A 3D object detection algorithm based on Lidar-camera fusion is designed to detect the target objects. Based on the Hungarian algorithm, this paper designs a 3D multi-object tracking algorithm with an adaptive threshold to realize the stable matching and tracking of the objects. We combine Memory Augmented Neural Networks (MANN) and Kalman filter to achieve 3D trajectory prediction on rugged terrains. Besides, we realize a new dynamic SLAM by using the results of multi-object tracking to remove dynamic points for better SLAM performance and static map. To verify the effectiveness of the proposed multi-object tracking and prediction system, several simulations and physical experiments are conducted. The results show that the proposed system can track dynamic objects and provide future trajectory and a more clean static map in real-time.
Semi-Supervised Learning via Weight-aware Distillation under Class Distribution Mismatch
Authors: Pan Du, Suyun Zhao, Zisen Sheng, Cuiping Li, Hong Chen
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Semi-Supervised Learning (SSL) under class distribution mismatch aims to tackle a challenging problem wherein unlabeled data contain lots of unknown categories unseen in the labeled ones. In such mismatch scenarios, traditional SSL suffers severe performance damage due to the harmful invasion of the instances with unknown categories into the target classifier. In this study, by strict mathematical reasoning, we reveal that the SSL error under class distribution mismatch is composed of pseudo-labeling error and invasion error, both of which jointly bound the SSL population risk. To alleviate the SSL error, we propose a robust SSL framework called Weight-Aware Distillation (WAD) that, by weights, selectively transfers knowledge beneficial to the target task from unsupervised contrastive representation to the target classifier. Specifically, WAD captures adaptive weights and high-quality pseudo labels to target instances by exploring point mutual information (PMI) in representation space to maximize the role of unlabeled data and filter unknown categories. Theoretically, we prove that WAD has a tight upper bound of population risk under class distribution mismatch. Experimentally, extensive results demonstrate that WAD outperforms five state-of-the-art SSL approaches and one standard baseline on two benchmark datasets, CIFAR10 and CIFAR100, and an artificial cross-dataset. The code is available at https://github.com/RUC-DWBI-ML/research/tree/main/WAD-master.
Integrated Image and Location Analysis for Wound Classification: A Deep Learning Approach
Abstract
The global burden of acute and chronic wounds presents a compelling case for enhancing wound classification methods, a vital step in diagnosing and determining optimal treatments. Recognizing this need, we introduce an innovative multi-modal network based on a deep convolutional neural network for categorizing wounds into four categories: diabetic, pressure, surgical, and venous ulcers. Our multi-modal network uses wound images and their corresponding body locations for more precise classification. A unique aspect of our methodology is incorporating a body map system that facilitates accurate wound location tagging, improving upon traditional wound image classification techniques. A distinctive feature of our approach is the integration of models such as VGG16, ResNet152, and EfficientNet within a novel architecture. This architecture includes elements like spatial and channel-wise Squeeze-and-Excitation modules, Axial Attention, and an Adaptive Gated Multi-Layer Perceptron, providing a robust foundation for classification. Our multi-modal network was trained and evaluated on two distinct datasets comprising relevant images and corresponding location information. Notably, our proposed network outperformed traditional methods, reaching an accuracy range of 74.79% to 100% for Region of Interest (ROI) without location classifications, 73.98% to 100% for ROI with location classifications, and 78.10% to 100% for whole image classifications. This marks a significant enhancement over previously reported performance metrics in the literature. Our results indicate the potential of our multi-modal network as an effective decision-support tool for wound image classification, paving the way for its application in various clinical contexts.
HashReID: Dynamic Network with Binary Codes for Efficient Person Re-identification
Authors: Kshitij Nikhal, Yujunrong Ma, Shuvra S. Bhattacharyya, Benjamin S. Riggan
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Biometric applications, such as person re-identification (ReID), are often deployed on energy constrained devices. While recent ReID methods prioritize high retrieval performance, they often come with large computational costs and high search time, rendering them less practical in real-world settings. In this work, we propose an input-adaptive network with multiple exit blocks, that can terminate computation early if the retrieval is straightforward or noisy, saving a lot of computation. To assess the complexity of the input, we introduce a temporal-based classifier driven by a new training strategy. Furthermore, we adopt a binary hash code generation approach instead of relying on continuous-valued features, which significantly improves the search process by a factor of 20. To ensure similarity preservation, we utilize a new ranking regularizer that bridges the gap between continuous and binary features. Extensive analysis of our proposed method is conducted on three datasets: Market1501, MSMT17 (Multi-Scene Multi-Time), and the BGC1 (BRIAR Government Collection). Using our approach, more than 70% of the samples with compact hash codes exit early on the Market1501 dataset, saving 80% of the networks computational cost and improving over other hash-based methods by 60%. These results demonstrate a significant improvement over dynamic networks and showcase comparable accuracy performance to conventional ReID methods. Code will be made available.
Edge-aware Hard Clustering Graph Pooling for Brain Imaging Data
Authors: Cheng Zhu, Jiayi Zhu, Lijuan Zhang, Xi Wu, Shuqi Yang, Ping Liang, Honghan Chen, Ying Tan
Subjects: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
Abstract
Graph Convolutional Networks (GCNs) can capture non-Euclidean spatial dependence between different brain regions, and the graph pooling operator in GCNs is key to enhancing the representation learning capability and acquiring abnormal brain maps. However, the majority of existing research designs graph pooling operators only from the perspective of nodes while disregarding the original edge features, in a way that not only confines graph pooling application scenarios, but also diminishes its ability to capture critical substructures. In this study, a clustering graph pooling method that first supports multidimensional edge features, called Edge-aware hard clustering graph pooling (EHCPool), is developed. EHCPool proposes the first 'Edge-to-node' score evaluation criterion based on edge features to assess node feature significance. To more effectively capture the critical subgraphs, a novel Iteration n-top strategy is further designed to adaptively learn sparse hard clustering assignments for graphs. Subsequently, an innovative N-E Aggregation strategy is presented to aggregate node and edge feature information in each independent subgraph. The proposed model was evaluated on multi-site brain imaging public datasets and yielded state-of-the-art performance. We believe this method is the first deep learning tool with the potential to probe different types of abnormal functional brain networks from data-driven perspective.
Addressing Selection Bias in Computerized Adaptive Testing: A User-Wise Aggregate Influence Function Approach
Authors: Soonwoo Kwon, Sojung Kim, Seunghyun Lee, Jin-Young Kim, Suyeong An, Kyuseok Kim
Subjects: Machine Learning (cs.LG); Computers and Society (cs.CY)
Abstract
Computerized Adaptive Testing (CAT) is a widely used, efficient test mode that adapts to the examinee's proficiency level in the test domain. CAT requires pre-trained item profiles, for CAT iteratively assesses the student real-time based on the registered items' profiles, and selects the next item to administer using candidate items' profiles. However, obtaining such item profiles is a costly process that involves gathering a large, dense item-response data, then training a diagnostic model on the collected data. In this paper, we explore the possibility of leveraging response data collected in the CAT service. We first show that this poses a unique challenge due to the inherent selection bias introduced by CAT, i.e., more proficient students will receive harder questions. Indeed, when naively training the diagnostic model using CAT response data, we observe that item profiles deviate significantly from the ground-truth. To tackle the selection bias issue, we propose the user-wise aggregate influence function method. Our intuition is to filter out users whose response data is heavily biased in an aggregate manner, as judged by how much perturbation the added data will introduce during parameter estimation. This way, we may enhance the performance of CAT while introducing minimal bias to the item profiles. We provide extensive experiments to demonstrate the superiority of our proposed method based on the three public datasets and one dataset that contains real-world CAT response data.
Adaptive Encoding Strategies for Erasing-Based Lossless Floating-Point Compression
Abstract
Lossless floating-point time series compression is crucial for a wide range of critical scenarios. Nevertheless, it is a big challenge to compress time series losslessly due to the complex underlying layouts of floating-point values. The state-of-the-art erasing-based compression algorithm Elf demonstrates a rather impressive performance. We give an in-depth exploration of the encoding strategies of Elf, and find that there is still much room for improvement. In this paper, we propose Elf, which employs a set of optimizations for leading zeros, center bits and sharing condition. Specifically, we develop a dynamic programming algorithm with a set of pruning strategies to compute the adaptive approximation rules efficiently. We theoretically prove that the adaptive approximation rules are globally optimal. We further extend Elf to Streaming Elf, i.e., SElf, which achieves almost the same compression ratio as Elf, while enjoying even higher efficiency in streaming scenarios. We compare Elf and SElf with 8 competitors using 22 datasets. The results demonstrate that SElf achieves 9.2% relative compression ratio improvement over the best streaming competitor while maintaining similar efficiency, and that Elf* ranks among the most competitive batch compressors. All source codes are publicly released.
OFVL-MS: Once for Visual Localization across Multiple Indoor Scenes
Authors: Tao Xie, Kun Dai, Siyi Lu, Ke Wang, Zhiqiang Jiang, Jinghan Gao, Dedong Liu, Jie Xu, Lijun Zhao, Ruifeng Li
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
In this work, we seek to predict camera poses across scenes with a multi-task learning manner, where we view the localization of each scene as a new task. We propose OFVL-MS, a unified framework that dispenses with the traditional practice of training a model for each individual scene and relieves gradient conflict induced by optimizing multiple scenes collectively, enabling efficient storage yet precise visual localization for all scenes. Technically, in the forward pass of OFVL-MS, we design a layer-adaptive sharing policy with a learnable score for each layer to automatically determine whether the layer is shared or not. Such sharing policy empowers us to acquire task-shared parameters for a reduction of storage cost and task-specific parameters for learning scene-related features to alleviate gradient conflict. In the backward pass of OFVL-MS, we introduce a gradient normalization algorithm that homogenizes the gradient magnitude of the task-shared parameters so that all tasks converge at the same pace. Furthermore, a sparse penalty loss is applied on the learnable scores to facilitate parameter sharing for all tasks without performance degradation. We conduct comprehensive experiments on multiple benchmarks and our new released indoor dataset LIVL, showing that OFVL-MS families significantly outperform the state-of-the-arts with fewer parameters. We also verify that OFVL-MS can generalize to a new scene with much few parameters while gaining superior localization performance.
Synergistic Multiscale Detail Refinement via Intrinsic Supervision for Underwater Image Enhancement
Abstract
Visual restoration of underwater scenes is crucial for visual tasks, and avoiding interference from underwater media has become a prominent concern. In this work, we present a synergistic multiscale detail refinement via intrinsic supervision (SMDR-IS) to recover underwater scene details. The low-degradation stage provides multiscale detail for original stage, which achieves synergistic multiscale detail refinement through feature propagation via the adaptive selective intrinsic supervised feature module (ASISF), which achieves synergistic multiscale detail refinement. ASISF is developed using intrinsic supervision to precisely control and guide feature transmission in the multi-degradation stages. ASISF improves the multiscale detail refinement while reducing interference from irrelevant scene information from the low-degradation stage. Additionally, within the multi-degradation encoder-decoder of SMDR-IS, we introduce a bifocal intrinsic-context attention module (BICA). This module is designed to effectively leverage multi-scale scene information found in images, using intrinsic supervision principles as its foundation. BICA facilitates the guidance of higher-resolution spaces by leveraging lower-resolution spaces, considering the significant dependency of underwater image restoration on spatial contextual relationships. During the training process, the network gains advantages from the integration of a multi-degradation loss function. This function serves as a constraint, enabling the network to effectively exploit information across various scales. When compared with state-of-the-art methods, SMDR-IS demonstrates its outstanding performance. Code will be made publicly available.
Boosting Diffusion Models with an Adaptive Momentum Sampler
Abstract
Diffusion probabilistic models (DPMs) have been shown to generate high-quality images without the need for delicate adversarial training. However, the current sampling process in DPMs is prone to violent shaking. In this paper, we present a novel reverse sampler for DPMs inspired by the widely-used Adam optimizer. Our proposed sampler can be readily applied to a pre-trained diffusion model, utilizing momentum mechanisms and adaptive updating to smooth the reverse sampling process and ensure stable generation, resulting in outputs of enhanced quality. By implicitly reusing update directions from early steps, our proposed sampler achieves a better balance between high-level semantics and low-level details. Additionally, this sampler is flexible and can be easily integrated into pre-trained DPMs regardless of the sampler used during training. Our experimental results on multiple benchmarks demonstrate that our proposed reverse sampler yields remarkable improvements over different baselines. We will make the source code available.
Efficient Transfer Learning in Diffusion Models via Adversarial Noise
Abstract
Diffusion Probabilistic Models (DPMs) have demonstrated substantial promise in image generation tasks but heavily rely on the availability of large amounts of training data. Previous works, like GANs, have tackled the limited data problem by transferring pre-trained models learned with sufficient data. However, those methods are hard to be utilized in DPMs since the distinct differences between DPM-based and GAN-based methods, showing in the unique iterative denoising process integral and the need for many timesteps with no-targeted noise in DPMs. In this paper, we propose a novel DPMs-based transfer learning method, TAN, to address the limited data problem. It includes two strategies: similarity-guided training, which boosts transfer with a classifier, and adversarial noise selection which adaptive chooses targeted noise based on the input image. Extensive experiments in the context of few-shot image generation tasks demonstrate that our method is not only efficient but also excels in terms of image quality and diversity when compared to existing GAN-based and DDPM-based methods.
Abstract
It is now possible to reconstruct dynamic human motion and shape from a sparse set of cameras using Neural Radiance Fields (NeRF) driven by an underlying skeleton. However, a challenge remains to model the deformation of cloth and skin in relation to skeleton pose. Unlike existing avatar models that are learned implicitly or rely on a proxy surface, our approach is motivated by the observation that different poses necessitate unique frequency assignments. Neglecting this distinction yields noisy artifacts in smooth areas or blurs fine-grained texture and shape details in sharp regions. We develop a two-branch neural network that is adaptive and explicit in the frequency domain. The first branch is a graph neural network that models correlations among body parts locally, taking skeleton pose as input. The second branch combines these correlation features to a set of global frequencies and then modulates the feature encoding. Our experiments demonstrate that our network outperforms state-of-the-art methods in terms of preserving details and generalization capabilities.
IncreLoRA: Incremental Parameter Allocation Method for Parameter-Efficient Fine-tuning
Abstract
With the increasing size of pre-trained language models (PLMs), fine-tuning all the parameters in the model is not efficient, especially when there are a large number of downstream tasks, which incur significant training and storage costs. Many parameter-efficient fine-tuning (PEFT) approaches have been proposed, among which, Low-Rank Adaptation (LoRA) is a representative approach that injects trainable rank decomposition matrices into every target module. Yet LoRA ignores the importance of parameters in different modules. To address this problem, many works have been proposed to prune the parameters of LoRA. However, under limited training conditions, the upper bound of the rank of the pruned parameter matrix is still affected by the preset values. We, therefore, propose IncreLoRA, an incremental parameter allocation method that adaptively adds trainable parameters during training based on the importance scores of each module. This approach is different from the pruning method as it is not limited by the initial number of training parameters, and each parameter matrix has a higher rank upper bound for the same training overhead. We conduct extensive experiments on GLUE to demonstrate the effectiveness of IncreLoRA. The results show that our method owns higher parameter efficiency, especially when under the low-resource settings where our method significantly outperforms the baselines. Our code is publicly available.
Towards Privacy-Supporting Fall Detection via Deep Unsupervised RGB2Depth Adaptation
Abstract
Fall detection is a vital task in health monitoring, as it allows the system to trigger an alert and therefore enabling faster interventions when a person experiences a fall. Although most previous approaches rely on standard RGB video data, such detailed appearance-aware monitoring poses significant privacy concerns. Depth sensors, on the other hand, are better at preserving privacy as they merely capture the distance of objects from the sensor or camera, omitting color and texture information. In this paper, we introduce a privacy-supporting solution that makes the RGB-trained model applicable in depth domain and utilizes depth data at test time for fall detection. To achieve cross-modal fall detection, we present an unsupervised RGB to Depth (RGB2Depth) cross-modal domain adaptation approach that leverages labelled RGB data and unlabelled depth data during training. Our proposed pipeline incorporates an intermediate domain module for feature bridging, modality adversarial loss for modality discrimination, classification loss for pseudo-labeled depth data and labeled source data, triplet loss that considers both source and target domains, and a novel adaptive loss weight adjustment method for improved coordination among various losses. Our approach achieves state-of-the-art results in the unsupervised RGB2Depth domain adaptation task for fall detection. Code is available at https://github.com/1015206533/privacy_supporting_fall_detection.
Trajectory Tracking Control of Dual-PAM Soft Actuator with Hysteresis Compensator
Abstract
Soft robotics is an emergent and swiftly evolving field. Pneumatic actuators are suitable for driving soft robots because of their superior performance. However, their control is not easy due to their hysteresis characteristics. In response to these challenges, we propose an adaptive control method to compensate hysteresis of a soft actuator. Employing a novel dual pneumatic artificial muscle (PAM) bending actuator, the innovative control strategy abates hysteresis effects by dynamically modulating gains within a traditional PID controller corresponding with the predicted motion of the reference trajectory. Through comparative experimental evaluation, we found that the new control method outperforms its conventional counterparts regarding tracking accuracy and response speed. Our work reveals a new direction for advancing control in soft actuators.
Cached Operator Reordering: A Unified View for Fast GNN Training
Authors: Julia Bazinska, Andrei Ivanov, Tal Ben-Nun, Nikoli Dryden, Maciej Besta, Siyuan Shen, Torsten Hoefler
Abstract
Graph Neural Networks (GNNs) are a powerful tool for handling structured graph data and addressing tasks such as node classification, graph classification, and clustering. However, the sparse nature of GNN computation poses new challenges for performance optimization compared to traditional deep neural networks. We address these challenges by providing a unified view of GNN computation, I/O, and memory. By analyzing the computational graphs of the Graph Convolutional Network (GCN) and Graph Attention (GAT) layers -- two widely used GNN layers -- we propose alternative computation strategies. We present adaptive operator reordering with caching, which achieves a speedup of up to 2.43x for GCN compared to the current state-of-the-art. Furthermore, an exploration of different caching schemes for GAT yields a speedup of up to 1.94x. The proposed optimizations save memory, are easily implemented across various hardware platforms, and have the potential to alleviate performance bottlenecks in training large-scale GNN models.
Mesh Conflation of Oblique Photogrammetric Models using Virtual Cameras and Truncated Signed Distance Field
Authors: Shuang Song, Rongjun Qin
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Conflating/stitching 2.5D raster digital surface models (DSM) into a large one has been a running practice in geoscience applications, however, conflating full-3D mesh models, such as those from oblique photogrammetry, is extremely challenging. In this letter, we propose a novel approach to address this challenge by conflating multiple full-3D oblique photogrammetric models into a single, and seamless mesh for high-resolution site modeling. Given two or more individually collected and created photogrammetric meshes, we first propose to create a virtual camera field (with a panoramic field of view) to incubate virtual spaces represented by Truncated Signed Distance Field (TSDF), an implicit volumetric field friendly for linear 3D fusion; then we adaptively leverage the truncated bound of meshes in TSDF to conflate them into a single and accurate full 3D site model. With drone-based 3D meshes, we show that our approach significantly improves upon traditional methods for model conflations, to drive new potentials to create excessively large and accurate full 3D mesh models in support of geoscience and environmental applications.
Survey of adaptive containerization architectures for HPC
Authors: Tiziano Müller, Nina Mujkanovic, Juan J. Durillo, Nicolay Hammer
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract
Containers offer an array of advantages that benefit research reproducibility and portability across groups and systems. As container tools mature, container security improves, and High-performance computing (HPC) and cloud system tools converge, supercomputing centers are increasingly integrating containers in their workflows. The technology selection process requires sufficient information on the diverse tools available, yet the majority of research into containers still focuses on cloud environments. We consider an adaptive containerization approach, with a focus on accelerating the deployment of applications and workflows on HPC systems using containers. To this end, we discuss the specific HPC requirements regarding container tools, and analyze the entire containerization stack, including container engines and registries, in-depth. Finally, we consider various orchestrator and HPC workload manager integration scenarios.
Model reference adaptive control for state and input constrained linear systems
Abstract
State and input constraints are ubiquitous in all engineering systems. In this article, we derive adaptive controllers for uncertain linear systems under pre-specified state and input constraints. Several modifications of the model reference adaptive control (MRAC) framework have been proposed to address input constraints in uncertain linear systems. Considering the infeasibility of arbitrary reference trajectories, reference modification has been implemented in the case of input constraints in literature. The resulting conditions on the reference and input signals are difficult to verify online. Similar results on state and input constraints together have also been proposed, albeit resulting in more complex and unverifiable conditions on the control. The primary objective of this article is therefore to account for state and input constraints in uncertain linear systems by providing easily verifiable conditions on the control and reference. A combination of reference modification and barrier Lyapunov methods in adaptive control are employed to arrive at these results.
SG-Former: Self-guided Transformer with Evolving Token Reallocation
Authors: Sucheng Ren, Xingyi Yang, Songhua Liu, Xinchao Wang
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Vision Transformer has demonstrated impressive success across various vision tasks. However, its heavy computation cost, which grows quadratically with respect to the token sequence length, largely limits its power in handling large feature maps. To alleviate the computation cost, previous works rely on either fine-grained self-attentions restricted to local small regions, or global self-attentions but to shorten the sequence length resulting in coarse granularity. In this paper, we propose a novel model, termed as Self-guided Transformer~(SG-Former), towards effective global self-attention with adaptive fine granularity. At the heart of our approach is to utilize a significance map, which is estimated through hybrid-scale self-attention and evolves itself during training, to reallocate tokens based on the significance of each region. Intuitively, we assign more tokens to the salient regions for achieving fine-grained attention, while allocating fewer tokens to the minor regions in exchange for efficiency and global receptive fields. The proposed SG-Former achieves performance superior to state of the art: our base size model achieves \textbf{84.7\%} Top-1 accuracy on ImageNet-1K, \textbf{51.2mAP} bbAP on CoCo, \textbf{52.7mIoU} on ADE20K surpassing the Swin Transformer by \textbf{+1.3\% / +2.7 mAP/ +3 mIoU}, with lower computation costs and fewer parameters. The code is available at \href{https://github.com/OliverRensu/SG-Former}{https://github.com/OliverRensu/SG-Former}
Multi-Objective Optimization for Sparse Deep Neural Network Training
Authors: S. S. Hotegni, S. Peitz, M. Berkemeier
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
Abstract
Different conflicting optimization criteria arise naturally in various Deep Learning scenarios. These can address different main tasks (i.e., in the setting of Multi-Task Learning), but also main and secondary tasks such as loss minimization versus sparsity. The usual approach is a simple weighting of the criteria, which formally only works in the convex setting. In this paper, we present a Multi-Objective Optimization algorithm using a modified Weighted Chebyshev scalarization for training Deep Neural Networks (DNNs) with respect to several tasks. By employing this scalarization technique, the algorithm can identify all optimal solutions of the original problem while reducing its complexity to a sequence of single-objective problems. The simplified problems are then solved using an Augmented Lagrangian method, enabling the use of popular optimization techniques such as Adam and Stochastic Gradient Descent, while efficaciously handling constraints. Our work aims to address the (economical and also ecological) sustainability issue of DNN models, with a particular focus on Deep Multi-Task models, which are typically designed with a very large number of weights to perform equally well on multiple tasks. Through experiments conducted on two Machine Learning datasets, we demonstrate the possibility of adaptively sparsifying the model during training without significantly impacting its performance, if we are willing to apply task-specific adaptations to the network weights. Code is available at https://github.com/salomonhotegni/MDMTN.
Keyword: quantization
Compressed Models Decompress Race Biases: What Quantized Models Forget for Fair Face Recognition
Authors: Pedro C. Neto, Eduarda Caldeira, Jaime S. Cardoso, Ana F. Sequeira
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
With the ever-growing complexity of deep learning models for face recognition, it becomes hard to deploy these systems in real life. Researchers have two options: 1) use smaller models; 2) compress their current models. Since the usage of smaller models might lead to concerning biases, compression gains relevance. However, compressing might be also responsible for an increase in the bias of the final model. We investigate the overall performance, the performance on each ethnicity subgroup and the racial bias of a State-of-the-Art quantization approach when used with synthetic and real data. This analysis provides a few more details on potential benefits of performing quantization with synthetic data, for instance, the reduction of biases on the majority of test scenarios. We tested five distinct architectures and three different training datasets. The models were evaluated on a fourth dataset which was collected to infer and compare the performance of face recognition models on different ethnicity.
Keyword: efficient
Software-based signal compression algorithm for ROM-stored electrical cables
Pseudo-online framework for BCI evaluation: A MOABB perspective
An engine to simulate insurance fraud network data
VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection
Efficient Benchmarking (of Language Models)
VBMO: Voting-Based Multi-Objective Path Planning
Weakly Supervised Face and Whole Body Recognition in Turbulent Environments
3ET: Efficient Event-based Eye Tracking using a Change-Based ConvLSTM Network
Understanding Hessian Alignment for Domain Generalization
Few-shot Anomaly Detection in Text with Deviation Learning
Resource Allocation in Cloud Computing Using Genetic Algorithm and Neural Network
Towards Safe Automated Refactoring of Imperative Deep Learning Programs to Graph Execution
HypBO: Expert-Guided Chemist-in-the-Loop Bayesian Search for New Materials
Karasu: A Collaborative Approach to Efficient Cluster Configuration for Big Data Analytics
Flexible Multi-DoF Aerial 3D Printing Supported with Automated Optimal Chunking
Integrated Image and Location Analysis for Wound Classification: A Deep Learning Approach
Cabrita: closing the gap for foreign languages
The macroscopic finite-difference scheme and modified equations of the general propagation multiple-relaxation-time lattice Boltzmann model
Addressing Selection Bias in Computerized Adaptive Testing: A User-Wise Aggregate Influence Function Approach
Adaptive Encoding Strategies for Erasing-Based Lossless Floating-Point Compression
LFS-GAN: Lifelong Few-Shot Image Generation
Diverse Policies Converge in Reward-free Markov Decision Processe
OFVL-MS: Once for Visual Localization across Multiple Indoor Scenes
System Identification for Continuous-time Linear Dynamical Systems
Efficient Transfer Learning in Diffusion Models via Adversarial Noise
CED: Consistent ensemble distillation for audio tagging
EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE
ESTA: An Efficient Spatial-Temporal Range Aggregation Query Processing Algorithm for UAV Networks
RankMixup: Ranking-Based Mixup Training for Network Calibration
StofNet: Super-resolution Time of Flight Network
Bias-Aware Minimisation: Understanding and Mitigating Estimator Bias in Private SGD
MARC: Multipolicy and Risk-aware Contingency Planning for Autonomous Driving
Reranking Passages with Coarse-to-Fine Neural Retriever using List-Context Information
IncreLoRA: Incremental Parameter Allocation Method for Parameter-Efficient Fine-tuning
A multiobjective continuation method to compute the regularization path of deep neural networks
Layer-wise Feedback Propagation
InstructionGPT-4: A 200-Instruction Paradigm for Fine-Tuning MiniGPT-4
State Merging with Quantifiers in Symbolic Execution
Using the TypeScript compiler to fix erroneous Node.js snippets
Path-Constrained State Estimation for Rail Vehicles
Out of the Cage: How Stochastic Parrots Win in Cyber Security Environments
Multi-UAV Deployment in Obstacle-Cluttered Environments with LOS Connectivity
DarkDiff: Explainable web page similarity of TOR onion sites
Modeling excitable cells with the EMI equations: spectral analysis and iterative solution strategy
Unsupervised anomalies detection in IIoT edge devices networks using federated learning
Curriculum Learning with Adam: The Devil Is in the Wrong Details
A Heuristic Informative-Path-Planning Algorithm for Autonomous Mapping of Unknown Areas
How to Protect Copyright Data in Optimization of Large Language Models?
Language Reward Modulation for Pretraining Reinforcement Learning
Keyword: faster
Utilizing Admissible Bounds for Heuristic Learning
EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE
Towards Privacy-Supporting Fall Detection via Deep Unsupervised RGB2Depth Adaptation
Keyword: mobile
Towards an On-device Agent for Text Rewriting
PatchBackdoor: Backdoor Attack against Deep Neural Networks without Model Modification
A deep reinforcement learning approach for real-time demand-responsive railway rescheduling to mitigate station overcrowding using mobile data
Value of Assistance for Mobile Agents
Keyword: pruning
Adaptive Encoding Strategies for Erasing-Based Lossless Floating-Point Compression
IncreLoRA: Incremental Parameter Allocation Method for Parameter-Efficient Fine-tuning
Keyword: diffusion
Revolutionizing TCAD Simulations with Universal Device Encoding and Graph Attention Networks
The macroscopic finite-difference scheme and modified equations of the general propagation multiple-relaxation-time lattice Boltzmann model
Shape-conditioned 3D Molecule Generation via Equivariant Diffusion Models
Boosting Diffusion Models with an Adaptive Momentum Sampler
LongDanceDiff: Long-term Dance Generation with Conditional Diffusion Model
Efficient Transfer Learning in Diffusion Models via Adversarial Noise
High-quality Image Dehazing with Diffusion Model
Manipulating Embeddings of Stable Diffusion Prompts
Space-time hybridizable discontinuous Galerkin method for advection-diffusion on deforming domains: The advection-dominated regime
Diffusion Language Models Can Perform Many Tasks with Scaling and Instruction-Finetuning
On-Manifold Projected Gradient Descent
Keyword: adaptive
Software-based signal compression algorithm for ROM-stored electrical cables
Reinforcement Learning -based Adaptation and Scheduling Methods for Multi-source DASH
A General Framework for Progressive Data Compression and Retrieval
Bayesian Learning for Dynamic Target Localization with Human-provided Spatial Information
SEA: Shareable and Explainable Attribution for Query-based Black-box Attacks
Multi-object Detection, Tracking and Prediction in Rugged Dynamic Environments
Semi-Supervised Learning via Weight-aware Distillation under Class Distribution Mismatch
Integrated Image and Location Analysis for Wound Classification: A Deep Learning Approach
HashReID: Dynamic Network with Binary Codes for Efficient Person Re-identification
Edge-aware Hard Clustering Graph Pooling for Brain Imaging Data
Addressing Selection Bias in Computerized Adaptive Testing: A User-Wise Aggregate Influence Function Approach
Adaptive Encoding Strategies for Erasing-Based Lossless Floating-Point Compression
OFVL-MS: Once for Visual Localization across Multiple Indoor Scenes
Synergistic Multiscale Detail Refinement via Intrinsic Supervision for Underwater Image Enhancement
Boosting Diffusion Models with an Adaptive Momentum Sampler
Efficient Transfer Learning in Diffusion Models via Adversarial Noise
Pose Modulated Avatars from Video
IncreLoRA: Incremental Parameter Allocation Method for Parameter-Efficient Fine-tuning
Towards Privacy-Supporting Fall Detection via Deep Unsupervised RGB2Depth Adaptation
Trajectory Tracking Control of Dual-PAM Soft Actuator with Hysteresis Compensator
Cached Operator Reordering: A Unified View for Fast GNN Training
Mesh Conflation of Oblique Photogrammetric Models using Virtual Cameras and Truncated Signed Distance Field
Survey of adaptive containerization architectures for HPC
Model reference adaptive control for state and input constrained linear systems
SG-Former: Self-guided Transformer with Evolving Token Reallocation
Multi-Objective Optimization for Sparse Deep Neural Network Training
Keyword: quantization
Compressed Models Decompress Race Biases: What Quantized Models Forget for Fair Face Recognition