Abstract
In the stochastic gradient descent (SGD) for sequential simulations such as the neural stochastic differential equations, the Multilevel Monte Carlo (MLMC) method is known to offer better theoretical computational complexity compared to the naive Monte Carlo approach. However, in practice, MLMC scales poorly on massively parallel computing platforms such as modern GPUs, because of its large parallel complexity which is equivalent to that of the naive Monte Carlo method. To cope with this issue, we propose the delayed MLMC gradient estimator that drastically reduces the parallel complexity of MLMC by recycling previously computed gradient components from earlier steps of SGD. The proposed estimator provably reduces the average parallel complexity per iteration at the cost of a slightly worse per-iteration convergence rate. In our numerical experiments, we use an example of deep hedging to demonstrate the superior parallel complexity of our method compared to the standard MLMC in SGD.
High-dimensional SGD aligns with emerging outlier eigenspaces
Abstract
We rigorously study the joint evolution of training dynamics via stochastic gradient descent (SGD) and the spectra of empirical Hessian and gradient matrices. We prove that in two canonical classification tasks for multi-class high-dimensional mixtures and either 1 or 2-layer neural networks, the SGD trajectory rapidly aligns with emerging low-rank outlier eigenspaces of the Hessian and gradient matrices. Moreover, in multi-layer settings this alignment occurs per layer, with the final layer's outlier eigenspace evolving over the course of training, and exhibiting rank deficiency when the SGD converges to sub-optimal classifiers. This establishes some of the rich predictions that have arisen from extensive numerical studies in the last decade about the spectra of Hessian and information matrices over the course of training in overparametrized networks.
Keyword: optimization
Online Proactive Multi-Task Assignment with Resource Availability Anticipation
Abstract
With the emergence of services and online applications as taxi dispatching, crowdsourcing, package or food delivery, industrials and researchers are paying attention to the online multi-task assignment optimization field to quickly and efficiently met demands. In this context, this paper is interested in the multi-task assignment problem where multiple requests (e.g. tasks) arrive over time and must be dynamically matched to (mobile) agents. This optimization problem is known to be NP-hard. In order to treat this problem with a proactive mindset, we propose to use a receding-horizon approach to determine which resources (e.g. taxis, mobile agents, drones, robots) would be available within this (possibly dynamic) receding-horizon to meet the current set of requests (i.e. tasks) as good as possible. Contrarily to several works in this domain, we have chosen to make no assumption concerning future locations of requests. To achieve fast optimized online solutions in terms of costs and amount of allocated tasks, we have designed a genetic algorithm based on a fitness function integrating the traveled distance and the age of the requests. We compared our proactive multi-task assignment with resource availability anticipation approach with a classical reactive approach. The results obtained in two benchmark problems, one synthetic and another based on real data, show that our resource availability anticipation method can achieve better results in terms of costs (e.g. traveled distance) and amount of allocated tasks than reactive approaches while decreasing resources idle time.
Reinforcement Learning from Automatic Feedback for High-Quality Unit Test Generation
Authors: Authors: Benjamin Steenhoek, Michele Tufano, Neel Sundaresan, Alexey Svyatkovskiy
Abstract
Software testing is a crucial aspect of software development, and the creation of high-quality tests that adhere to best practices is essential for effective maintenance. Recently, Large Language Models (LLMs) have gained popularity for code generation, including the automated creation of test cases. However, these LLMs are often trained on vast amounts of publicly available code, which may include test cases that do not adhere to best practices and may even contain test smells (anti-patterns). To address this issue, we propose a novel technique called Reinforcement Learning from Static Quality Metrics (RLSQM). To begin, we analyze the anti-patterns generated by the LLM and show that LLMs can generate undesirable test smells. Thus, we train specific reward models for each static quality metric, then utilize Proximal Policy Optimization (PPO) to train models for optimizing a single quality metric at a time. Furthermore, we amalgamate these rewards into a unified reward model aimed at capturing different best practices and quality aspects of tests. By comparing RL-trained models with those trained using supervised learning, we provide insights into how reliably utilize RL to improve test generation quality and into the effects of various training strategies. Our experimental results demonstrate that the RL-optimized model consistently generated high-quality test cases compared to the base LLM, improving the model by up to 21%, and successfully generates nearly 100% syntactically correct code. RLSQM also outperformed GPT-4 on four out of seven metrics. This represents a significant step towards enhancing the overall efficiency and reliability of software testing through Reinforcement Learning and static quality metrics. Our data are available at this link: https://figshare.com/s/ded476c8d4c221222849.
Computing a Sparse Approximate Inverse on Quantum Annealing Machines
Abstract
Many engineering problems involve solving large linear systems of equations. Conjugate gradient (CG) is one of the most popular iterative methods for solving such systems. However, CG typically requires a good preconditioner to speed up convergence. One such preconditioner is the sparse approximate inverse (SPAI). In this paper, we explore the computation of an SPAI on quantum annealing machines by solving a series of quadratic unconstrained binary optimization (QUBO) problems. Numerical experiments are conducted using both well-conditioned and poorly-conditioned linear systems arising from a 2D finite difference formulation of the Poisson problem.
Learning Diverse Skills for Local Navigation under Multi-constraint Optimality
Authors: Authors: Jin Cheng, Marin Vlastelica, Pavel Kolev, Chenhao Li, Georg Martius
Abstract
Despite many successful applications of data-driven control in robotics, extracting meaningful diverse behaviors remains a challenge. Typically, task performance needs to be compromised in order to achieve diversity. In many scenarios, task requirements are specified as a multitude of reward terms, each requiring a different trade-off. In this work, we take a constrained optimization viewpoint on the quality-diversity trade-off and show that we can obtain diverse policies while imposing constraints on their value functions which are defined through distinct rewards. In line with previous work, further control of the diversity level can be achieved through an attract-repel reward term motivated by the Van der Waals force. We demonstrate the effectiveness of our method on a local navigation task where a quadruped robot needs to reach the target within a finite horizon. Finally, our trained policies transfer well to the real 12-DoF quadruped robot, Solo12, and exhibit diverse agile behaviors with successful obstacle traversal.
GenCO: Generating Diverse Solutions to Design Problems with Combinatorial Nature
Abstract
Generating diverse objects (e.g., images) using generative models (such as GAN or VAE) has achieved impressive results in the recent years, to help solve many design problems that are traditionally done by humans. Going beyond image generation, we aim to find solutions to more general design problems, in which both the diversity of the design and conformity of constraints are important. Such a setting has applications in computer graphics, animation, industrial design, material science, etc, in which we may want the output of the generator to follow discrete/combinatorial constraints and penalize any deviation, which is non-trivial with existing generative models and optimization solvers. To address this, we propose GenCO, a novel framework that conducts end-to-end training of deep generative models integrated with embedded combinatorial solvers, aiming to uncover high-quality solutions aligned with nonlinear objectives. While structurally akin to conventional generative models, GenCO diverges in its role - it focuses on generating instances of combinatorial optimization problems rather than final objects (e.g., images). This shift allows finer control over the generated outputs, enabling assessments of their feasibility and introducing an additional combinatorial loss component. We demonstrate the effectiveness of our approach on a variety of generative tasks characterized by combinatorial intricacies, including game level generation and map creation for path planning, consistently demonstrating its capability to yield diverse, high-quality solutions that reliably adhere to user-specified combinatorial properties.
Distributionally Safe Reinforcement Learning under Model Uncertainty: A Single-Level Approach by Differentiable Convex Programming
Authors: Authors: Alaa Eddine Chriat, Chuangchuang Sun
Subjects: Machine Learning (cs.LG); Robotics (cs.RO); Systems and Control (eess.SY)
Abstract
Safety assurance is uncompromisable for safety-critical environments with the presence of drastic model uncertainties (e.g., distributional shift), especially with humans in the loop. However, incorporating uncertainty in safe learning will naturally lead to a bi-level problem, where at the lower level the (worst-case) safety constraint is evaluated within the uncertainty ambiguity set. In this paper, we present a tractable distributionally safe reinforcement learning framework to enforce safety under a distributional shift measured by a Wasserstein metric. To improve the tractability, we first use duality theory to transform the lower-level optimization from infinite-dimensional probability space where distributional shift is measured, to a finite-dimensional parametric space. Moreover, by differentiable convex programming, the bi-level safe learning problem is further reduced to a single-level one with two sequential computationally efficient modules: a convex quadratic program to guarantee safety followed by a projected gradient ascent to simultaneously find the worst-case uncertainty. This end-to-end differentiable framework with safety constraints, to the best of our knowledge, is the first tractable single-level solution to address distributional safety. We test our approach on first and second-order systems with varying complexities and compare our results with the uncertainty-agnostic policies, where our approach demonstrates a significant improvement on safety guarantees.
SGORP: A Subgradient-based Method for d-Dimensional Rectilinear Partitioning
Authors: Authors: Muhammed Fatih Balin, Xiaojing An, Abdurrahman Yaşar, Ümit V. Çatalyürek
Abstract
Partitioning for load balancing is a crucial first step to parallelize any type of computation. In this work, we propose SGORP, a new spatial partitioning method based on Subgradient Optimization, to solve the $d$-dimensional Rectilinear Partitioning Problem (RPP). Our proposed method allows the use of customizable objective functions as well as some user-specific constraints, such as symmetric partitioning on selected dimensions. Extensive experimental evaluation using over 600 test matrices shows that our algorithm achieves favorable performance against the state-of-the-art RPP and Symmetric RPP algorithms. Additionally, we show the effectiveness of our algorithm to do application-specific load balancing using two applications as motivation: Triangle Counting and Sparse Matrix Multiplication (SpGEMM), where we model their load-balancing problems as $3$-dimensional RPPs.
A Recipe for Improved Certifiable Robustness: Capacity and Data
Authors: Authors: Kai Hu, Klas Leino, Zifan Wang, Matt Fredrikson
Abstract
A key challenge, supported both theoretically and empirically, is that robustness demands greater network capacity and more data than standard training. However, effectively adding capacity under stringent Lipschitz constraints has proven more difficult than it may seem, evident by the fact that state-of-the-art approach tend more towards \emph{underfitting} than overfitting. Moreover, we posit that a lack of careful exploration of the design space for Lipshitz-based approaches has left potential performance gains on the table. In this work, we provide a more comprehensive evaluation to better uncover the potential of Lipschitz-based certification methods. Using a combination of novel techniques, design optimizations, and synthesis of prior work, we are able to significantly improve the state-of-the-art \emph{verified robust accuracy} (VRA) for deterministic certification on a variety of benchmark datasets, and over a range of perturbation sizes. Of particular note, we discover that the addition of large "Cholesky-orthogonalized residual dense" layers to the end of existing state-of-the-art Lipschitz-controlled ResNet architectures is especially effective for increasing network capacity and performance. Combined with filtered generative data augmentation, our final results further the state of the art deterministic VRA by up to 8.5 percentage points. Code is available at \url{https://github.com/hukkai/liresnet}.
Parameterized Convex Minorant for Objective Function Approximation in Amortized Optimization
Authors: Authors: Jinrae Kim, Youdan Kim
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC)
Abstract
Parameterized convex minorant (PCM) method is proposed for the approximation of the objective function in amortized optimization. In the proposed method, the objective function approximator is expressed by the sum of a PCM and a nonnegative gap function, where the objective function approximator is bounded from below by the PCM convex in the optimization variable. The proposed objective function approximator is a universal approximator for continuous functions, and the global minimizer of the PCM attains the global minimum of the objective function approximator. Therefore, the global minimizer of the objective function approximator can be obtained by a single convex optimization. As a realization of the proposed method, extended parameterized log-sum-exp network is proposed by utilizing a parameterized log-sum-exp network as the PCM. Numerical simulation is performed for non-parameterized-convex objective function approximation and for learning-based nonlinear model predictive control to demonstrate the performance and characteristics of the proposed method. The simulation results support that the proposed method can be used to learn objective functions and to find the global minimizer reliably and quickly by using convex optimization algorithms.
Abstract
Conditional stochastic optimization has found applications in a wide range of machine learning tasks, such as invariant learning, AUPRC maximization, and meta-learning. As the demand for training models with large-scale distributed data grows in these applications, there is an increasing need for communication-efficient distributed optimization algorithms, such as federated learning algorithms. This paper considers the nonconvex conditional stochastic optimization in federated learning and proposes the first federated conditional stochastic optimization algorithm (FCSG) with a conditional stochastic gradient estimator and a momentum-based algorithm (FCSG-M). To match the lower bound complexity in the single-machine setting, we design an accelerated algorithm (Acc-FCSG-M) via the variance reduction to achieve the best sample and communication complexity. Compared with the existing optimization analysis for MAML in FL, federated conditional stochastic optimization considers the sample of tasks. Extensive experimental results on various tasks validate the efficiency of these algorithms.
Auto-FP: An Experimental Study of Automated Feature Preprocessing for Tabular Data
Authors: Authors: Danrui Qi, Jinglin Peng, Yongjun He, Jiannan Wang
Abstract
Classical machine learning models, such as linear models and tree-based models, are widely used in industry. These models are sensitive to data distribution, thus feature preprocessing, which transforms features from one distribution to another, is a crucial step to ensure good model quality. Manually constructing a feature preprocessing pipeline is challenging because data scientists need to make difficult decisions about which preprocessors to select and in which order to compose them. In this paper, we study how to automate feature preprocessing (Auto-FP) for tabular data. Due to the large search space, a brute-force solution is prohibitively expensive. To address this challenge, we interestingly observe that Auto-FP can be modelled as either a hyperparameter optimization (HPO) or a neural architecture search (NAS) problem. This observation enables us to extend a variety of HPO and NAS algorithms to solve the Auto-FP problem. We conduct a comprehensive evaluation and analysis of 15 algorithms on 45 public ML datasets. Overall, evolution-based algorithms show the leading average ranking. Surprisingly, the random search turns out to be a strong baseline. Many surrogate-model-based and bandit-based search algorithms, which achieve good performance for HPO and NAS, do not outperform random search for Auto-FP. We analyze the reasons for our findings and conduct a bottleneck analysis to identify the opportunities to improve these algorithms. Furthermore, we explore how to extend Auto-FP to support parameter search and compare two ways to achieve this goal. In the end, we evaluate Auto-FP in an AutoML context and discuss the limitations of popular AutoML tools. To the best of our knowledge, this is the first study on automated feature preprocessing. We hope our work can inspire researchers to develop new algorithms tailored for Auto-FP.
Tightly Joining Positioning and Control for Trustworthy Unmanned Aerial Vehicles Based on Factor Graph Optimization in Urban Transportation
Abstract
Unmanned aerial vehicles (UAV) showed great potential in improving the efficiency of parcel delivery applications in the coming smart cities era. Unfortunately, the trustworthy positioning and control algorithms of the UAV are significantly challenged in complex urban areas. For example, the ubiquitous global navigation satellite system (GNSS) positioning can be degraded by the signal reflections from surrounding high-rising buildings, leading to significantly increased positioning uncertainty. An additional challenge is introduced to the control algorithm due to the complex wind disturbances in urban canyons. Given the fact that the system positioning and control are highly correlated with each other, for example, the system dynamics of the control can largely help with the positioning, this paper proposed a joint positioning and control method (JPCM) based on factor graph optimization (FGO), which combines sensors' measurements and control intention. In particular, the positioning measurements are formulated as the factors in the factor graph model, such as the positioning from the GNSS. The model predictive control (MPC) is also formulated as the additional factors in the factor graph model. By solving the factor graph contributed by both the positioning factor and the MPC-based factors, the complementariness of positioning and control can be fully explored. To guarantee reliable system dynamic parameters, we validate the effectiveness of the proposed method using a simulated quadrotor system which showed significantly improved trajectory following performance. To benefit the research community, we open-source our code and make it available at https://github.com/RoboticsPolyu/IPN_MPC.
A ModelOps-based Framework for Intelligent Medical Knowledge Extraction
Abstract
Extracting medical knowledge from healthcare texts enhances downstream tasks like medical knowledge graph construction and clinical decision-making. However, the construction and application of knowledge extraction models lack automation, reusability and unified management, leading to inefficiencies for researchers and high barriers for non-AI experts such as doctors, to utilize knowledge extraction. To address these issues, we propose a ModelOps-based intelligent medical knowledge extraction framework that offers a low-code system for model selection, training, evaluation and optimization. Specifically, the framework includes a dataset abstraction mechanism based on multi-layer callback functions, a reusable model training, monitoring and management mechanism. We also propose a model recommendation method based on dataset similarity, which helps users quickly find potentially suitable models for a given dataset. Our framework provides convenience for researchers to develop models and simplifies model access for non-AI experts such as doctors.
Adaptive Spatio-Temporal Voxels Based Trajectory Planning for Autonomous Driving in Highway Traffic Flow
Abstract
Trajectory planning is crucial for the safe driving of autonomous vehicles in highway traffic flow. Currently, some advanced trajectory planning methods utilize spatio-temporal voxels to construct feasible regions and then convert trajectory planning into optimization problem solving based on the feasible regions. However, these feasible region construction methods cannot adapt to the changes in dynamic environments, making them difficult to apply in complex traffic flow. In this paper, we propose a trajectory planning method based on adaptive spatio-temporal voxels which improves the construction of feasible regions and trajectory optimization while maintaining the quadratic programming form. The method can adjust feasible regions and trajectory planning according to real-time traffic flow and environmental changes, realizing vehicles to drive safely in complex traffic flow. The proposed method has been tested in both open-loop and closed-loop environments, and the test results show that our method outperforms the current planning methods.
Long-Term Dynamic Window Approach for Kinodynamic Local Planning in Static and Crowd Environments
Abstract
Local planning for a differential wheeled robot is designed to generate kinodynamic feasible actions that guide the robot to a goal position along the navigation path while avoiding obstacles. Reactive, predictive, and learning-based methods are widely used in local planning. However, few of them can fit static and crowd environments while satisfying kinodynamic constraints simultaneously. To solve this problem, we propose a novel local planning method. The method applies a long-term dynamic window approach to generate an initial trajectory and then optimizes it with graph optimization. The method can plan actions under the robot's kinodynamic constraints in real time while allowing the generated actions to be safer and more jitterless. Experimental results show that the proposed method adapts well to crowd and static environments and outperforms most SOTA approaches.
Exploring Federated Optimization by Reducing Variance of Adaptive Unbiased Client Sampling
Abstract
Federated Learning (FL) systems usually sample a fraction of clients to conduct a training process. Notably, the variance of global estimates for updating the global model built on information from sampled clients is highly related to federated optimization quality. This paper explores a line of "free" adaptive client sampling techniques in federated optimization, where the server builds promising sampling probability and reliable global estimates without requiring additional local communication and computation. We capture a minor variant in the sampling procedure and improve the global estimation accordingly. Based on that, we propose a novel sampler called K-Vib, which solves an online convex optimization respecting client sampling in federated optimization. It achieves improved a linear speed up on regret bound $\tilde{\mathcal{O}}\big(N^{\frac{1}{3}}T^{\frac{2}{3}}/K^{\frac{4}{3}}\big)$ with communication budget $K$. As a result, it significantly improves the performance of federated optimization. Theoretical improvements and intensive experiments on classic federated tasks demonstrate our findings.
Tackling Hybrid Heterogeneity on Federated Optimization via Gradient Diversity Maximization
Abstract
Federated learning refers to a distributed machine learning paradigm in which data samples are decentralized and distributed among multiple clients. These samples may exhibit statistical heterogeneity, which refers to data distributions are not independent and identical across clients. Additionally, system heterogeneity, or variations in the computational power of the clients, introduces biases into federated learning. The combined effects of statistical and system heterogeneity can significantly reduce the efficiency of federated optimization. However, the impact of hybrid heterogeneity is not rigorously discussed. This paper explores how hybrid heterogeneity affects federated optimization by investigating server-side optimization. The theoretical results indicate that adaptively maximizing gradient diversity in server update direction can help mitigate the potential negative consequences of hybrid heterogeneity. To this end, we introduce a novel server-side gradient-based optimizer \textsc{FedAWARE} with theoretical guarantees provided. Intensive experiments in heterogeneous federated settings demonstrate that our proposed optimizer can significantly enhance the performance of federated learning across varying degrees of hybrid heterogeneity.
Understanding Pan-Sharpening via Generalized Inverse
Authors: Authors: Shiqi Liu, Yutong Bai, Xinyang Han, Alan Yuille
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Pan-sharpening algorithm utilizes panchromatic image and multispectral image to obtain a high spatial and high spectral image. However, the optimizations of the algorithms are designed with different standards. We adopt the simple matrix equation to describe the Pan-sharpening problem. The solution existence condition and the acquirement of spectral and spatial resolution are discussed. A down-sampling enhancement method was introduced for better acquiring the spatial and spectral down-sample matrices. By the generalized inverse theory, we derived two forms of general inverse matrix formulations that can correspond to the two prominent classes of Pan-sharpening methods, that is, component substitution and multi-resolution analysis methods. Specifically, the Gram Schmidt Adaptive(GSA) was proved to follow the general inverse matrix formulation of component substitution. A model prior to the general inverse matrix of the spectral function was rendered. The theoretical errors are analyzed. Synthetic experiments and real data experiments are implemented. The proposed methods are better and sharper than other methods qualitatively in both synthetic and real experiments. The down-sample enhancement effect is shown of better results both quantitatively and qualitatively in real experiments. The generalized inverse matrix theory help us better understand the Pan-sharpening.
Efficient Quantification and Representation of Aggregate Flexibility in Electric Vehicles
Authors: Authors: Nanda Kishor Panda, Simon H. Tindemans
Abstract
Aggregation is crucial to the effective use of flexibility, especially in the case of electric vehicles (EVs) because of their limited individual battery sizes and large aggregate impact. This research proposes a novel method of quantifying and representing the aggregate flexibility of EV fleets within a fixed flexibility request window. These windows can be chosen based on relevant network operator needs, such as evening congestion periods. The proposed UL-flexibility is independent of the number of assets but scales only with the number of discrete time steps in the chosen window. The representation involves $2T$ parameters, with T being the number of time steps in the window. Feasibility of signals can be checked using $2T$ constraints and optimization using $2(2^T-1)$ constraints, both exactly capturing the flexibility region. Using a request window eliminates uncertainty related to EV arrival and departure times outside the window. We present the necessary theoretical framework for our proposed methods and outline steps for transitioning between representations. Additionally, we compare the computational efficiency of the proposed method with the common direct aggregation method.
Reward Model Ensembles Help Mitigate Overoptimization
Authors: Authors: Thomas Coste, Usman Anwar, Robert Kirk, David Krueger
Abstract
Reinforcement learning from human feedback (RLHF) is a standard approach for fine-tuning large language models to follow instructions. As part of this process, learned reward models are used to approximately model human preferences. However, as imperfect representations of the "true" reward, these learned reward models are susceptible to \textit{overoptimization}. Gao et al. (2023) studied this phenomenon in a synthetic human feedback setup with a significantly larger "gold" reward model acting as the true reward (instead of humans) and showed that overoptimization remains a persistent problem regardless of the size of the proxy reward model and training data used. Using a similar setup, we conduct a systematic study to evaluate the efficacy of using ensemble-based conservative optimization objectives, specifically worst-case optimization (WCO) and uncertainty-weighted optimization (UWO), for mitigating reward model overoptimization when using two optimization methods: (a) best-of-n sampling (BoN) (b) proximal policy optimization (PPO). We additionally extend the setup of Gao et al. (2023) to include 25% label noise to better mirror real-world conditions. Both with and without label noise, we find that conservative optimization practically eliminates overoptimization and improves performance by up to 70% for BoN sampling. For PPO, ensemble-based conservative optimization always reduces overoptimization and outperforms single reward model optimization. Moreover, combining it with a small KL penalty successfully prevents overoptimization at no performance cost. Overall, our results demonstrate that ensemble-based conservative optimization can effectively counter overoptimization.
SHOT: Suppressing the Hessian along the Optimization Trajectory for Gradient-Based Meta-Learning
Abstract
In this paper, we hypothesize that gradient-based meta-learning (GBML) implicitly suppresses the Hessian along the optimization trajectory in the inner loop. Based on this hypothesis, we introduce an algorithm called SHOT (Suppressing the Hessian along the Optimization Trajectory) that minimizes the distance between the parameters of the target and reference models to suppress the Hessian in the inner loop. Despite dealing with high-order terms, SHOT does not increase the computational complexity of the baseline model much. It is agnostic to both the algorithm and architecture used in GBML, making it highly versatile and applicable to any GBML baseline. To validate the effectiveness of SHOT, we conduct empirical tests on standard few-shot learning tasks and qualitatively analyze its dynamics. We confirm our hypothesis empirically and demonstrate that SHOT outperforms the corresponding baseline. Code is available at: https://github.com/JunHoo-Lee/SHOT
R-LGP: A Reachability-guided Logic-geometric Programming Framework for Optimal Task and Motion Planning on Mobile Manipulators
Authors: Authors: Kim Tien Ly, Valeriy Semenov, Mattia Risiglione, Wolfgang Merkt, Ioannis Havoutis
Abstract
This paper presents an optimization-based solution to task and motion planning (TAMP) on mobile manipulators. Logic-geometric programming (LGP) has shown promising capabilities for optimally dealing with hybrid TAMP problems that involve abstract and geometric constraints. However, LGP does not scale well to high-dimensional systems (e.g. mobile manipulators) and can suffer from obstacle avoidance issues. In this work, we extend LGP with a sampling-based reachability graph to enable solving optimal TAMP on high-DoF mobile manipulators. The proposed reachability graph can incorporate environmental information (obstacles) to provide the planner with sufficient geometric constraints. This reachability-aware heuristic efficiently prunes infeasible sequences of actions in the continuous domain, hence, it reduces replanning by securing feasibility at the final full trajectory optimization. Our framework proves to be time-efficient in computing optimal and collision-free solutions, while outperforming the current state of the art on metrics of success rate, planning time, path length and number of steps. We validate our framework on the physical Toyota HSR robot and report comparisons on a series of mobile manipulation tasks of increasing difficulty.
Everest: GPU-Accelerated System For Mining Temporal Motifs
Abstract
Temporal motif mining is the task of finding the occurrences of subgraph patterns within a large input temporal graph that obey the specified structural and temporal constraints. Despite its utility in several critical application domains that demand high performance (e.g., detecting fraud in financial transaction graphs), the performance of existing software is limited on commercial hardware platforms, in that it runs for tens of hours. This paper presents Everest - a system that efficiently maps the workload of mining (supports both enumeration and counting) temporal motifs to the highly parallel GPU architecture. In particular, using an input temporal graph and a more expressive user-defined temporal motif query definition compared to prior works, Everest generates an execution plan and runtime primitives that optimize the workload execution by exploiting the high compute throughput of a GPU. Everest generates motif-specific mining code to reduce long-latency memory accesses and frequent thread divergence operations. Everest incorporates novel low-cost runtime mechanisms to enable load balancing to improve GPU hardware utilization. To support large graphs that do not fit on GPU memory, Everest also supports multi-GPU execution by intelligently partitioning the edge list that prevents inter-GPU communication. Everest hides the implementation complexity of presented optimizations away from the targeted system user for better usability. Our evaluation shows that, using proposed optimizations, Everest improves the performance of a baseline GPU implementation by 19x, on average.
A Deep Instance Generative Framework for MILP Solvers Under Limited Data Availability
Abstract
In the past few years, there has been an explosive surge in the use of machine learning (ML) techniques to address combinatorial optimization (CO) problems, especially mixed-integer linear programs (MILPs). Despite the achievements, the limited availability of real-world instances often leads to sub-optimal decisions and biased solver assessments, which motivates a suite of synthetic MILP instance generation techniques. However, existing methods either rely heavily on expert-designed formulations or struggle to capture the rich features of real-world instances. To tackle this problem, we propose G2MILP, which to the best of our knowledge is the first deep generative framework for MILP instances. Specifically, G2MILP represents MILP instances as bipartite graphs, and applies a masked variational autoencoder to iteratively corrupt and replace parts of the original graphs to generate new ones. The appealing feature of G2MILP is that it can learn to generate novel and realistic MILP instances without prior expert-designed formulations, while preserving the structures and computational hardness of real-world datasets, simultaneously. Thus the generated instances can facilitate downstream tasks for enhancing MILP solvers under limited data availability. We design a suite of benchmarks to evaluate the quality of the generated MILP instances. Experiments demonstrate that our method can produce instances that closely resemble real-world datasets in terms of both structures and computational hardness.
Incorporating Target Vehicle Trajectories Predicted by Deep Learning Into Model Predictive Controlled Vehicles
Authors: Authors: Ni Dang, Zengjie Zhang, Jizheng Liu, Marion Leibold, Martin Buss
Abstract
Model Predictive Control (MPC) has been widely applied to the motion planning of autonomous vehicles. An MPC-controlled vehicle is required to predict its own trajectories in a finite prediction horizon according to its model. Beyond this, the vehicle should also incorporate the prediction of the trajectory of its nearby vehicles, or target vehicles (TVs) into its decision-making. The conventional trajectory prediction methods, such as the constant-speed-based ones, are too trivial to accurately capture the potential collision risks. In this report, we propose a novel MPC-based motion planning method for an autonomous vehicle with a set of risk-aware constraints. These constraints incorporate the predicted trajectory of a TV learned using a deep-learning-based method. A recurrent neural network (RNN) is used to predict the TV's future trajectory based on its historical data. Then, the predicted TV trajectory is incorporated into the optimization of the MPC of the ego vehicle to generate collision-free motion. Simulation studies are conducted to showcase the prediction accuracy of the RNN model and the collision-free trajectories generated by the MPC.
Magicremover: Tuning-free Text-guided Image inpainting with Diffusion Models
Authors: Authors: Siyuan Yang, Lu Zhang, Liqian Ma, Yu Liu, JingJing Fu, You He
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Image inpainting aims to fill in the missing pixels with visually coherent and semantically plausible content. Despite the great progress brought from deep generative models, this task still suffers from i. the difficulties in large-scale realistic data collection and costly model training; and ii. the intrinsic limitations in the traditionally user-defined binary masks on objects with unclear boundaries or transparent texture. In this paper, we propose MagicRemover, a tuning-free method that leverages the powerful diffusion models for text-guided image inpainting. We introduce an attention guidance strategy to constrain the sampling process of diffusion models, enabling the erasing of instructed areas and the restoration of occluded content. We further propose a classifier optimization algorithm to facilitate the denoising stability within less sampling steps. Extensive comparisons are conducted among our MagicRemover and state-of-the-art methods including quantitative evaluation and user study, demonstrating the significant improvement of MagicRemover on high-quality image inpainting. We will release our code at https://github.com/exisas/Magicremover.
Approximating Robot Configuration Spaces with few Convex Sets using Clique Covers of Visibility Graphs
Authors: Authors: Peter Werner, Alexandre Amice, Tobia Marcucci, Daniela Rus, Russ Tedrake
Abstract
Many computations in robotics can be dramatically accelerated if the robot configuration space is described as a collection of simple sets. For example, recently developed motion planners rely on a convex decomposition of the free space to design collision-free trajectories using fast convex optimization. In this work, we present an efficient method for approximately covering complex configuration spaces with a small number of polytopes. The approach constructs a visibility graph using sampling and generates a clique cover of this graph to find clusters of samples that have mutual line of sight. These clusters are then inflated into large, full-dimensional, polytopes. We evaluate our method on a variety of robotic systems and show that it consistently covers larger portions of free configuration space, with fewer polytopes, and in a fraction of the time compared to previous methods.
A Grammatical Compositional Model for Video Action Detection
Abstract
Analysis of human actions in videos demands understanding complex human dynamics, as well as the interaction between actors and context. However, these interaction relationships usually exhibit large intra-class variations from diverse human poses or object manipulations, and fine-grained inter-class differences between similar actions. Thus the performance of existing methods is severely limited. Motivated by the observation that interactive actions can be decomposed into actor dynamics and participating objects or humans, we propose to investigate the composite property of them. In this paper, we present a novel Grammatical Compositional Model (GCM) for action detection based on typical And-Or graphs. Our model exploits the intrinsic structures and latent relationships of actions in a hierarchical manner to harness both the compositionality of grammar models and the capability of expressing rich features of DNNs. The proposed model can be readily embodied into a neural network module for efficient optimization in an end-to-end manner. Extensive experiments are conducted on the AVA dataset and the Something-Else task to demonstrate the superiority of our model, meanwhile the interpretability is enhanced through an inference parsing procedure.
CoLiDE: Concomitant Linear DAG Estimation
Authors: Authors: Seyed Saman Saboksayr, Gonzalo Mateos, Mariano Tepper
Abstract
We deal with the combinatorial problem of learning directed acyclic graph (DAG) structure from observational data adhering to a linear structural equation model (SEM). Leveraging advances in differentiable, nonconvex characterizations of acyclicity, recent efforts have advocated a continuous constrained optimization paradigm to efficiently explore the space of DAGs. Most existing methods employ lasso-type score functions to guide this search, which (i) require expensive penalty parameter retuning when the $\textit{unknown}$ SEM noise variances change across problem instances; and (ii) implicitly rely on limiting homoscedasticity assumptions. In this work, we propose a new convex score function for sparsity-aware learning of linear DAGs, which incorporates concomitant estimation of scale and thus effectively decouples the sparsity parameter from the exogenous noise levels. Regularization via a smooth, nonconvex acyclicity penalty term yields CoLiDE ($\textbf{Co}$ncomitant $\textbf{Li}$near $\textbf{D}$AG $\textbf{E}$stimation), a regression-based criterion amenable to efficient gradient computation and closed-form estimation of noise variances in heteroscedastic scenarios. Our algorithm outperforms state-of-the-art methods without incurring added complexity, especially when the DAGs are larger and the noise level profile is heterogeneous. We also find CoLiDE exhibits enhanced stability manifested via reduced standard deviations in several domain-specific metrics, underscoring the robustness of our novel linear DAG estimator.
Recovery of Training Data from Overparameterized Autoencoders: An Inverse Problem Perspective
Abstract
We study the recovery of training data from overparameterized autoencoder models. Given a degraded training sample, we define the recovery of the original sample as an inverse problem and formulate it as an optimization task. In our inverse problem, we use the trained autoencoder to implicitly define a regularizer for the particular training dataset that we aim to retrieve from. We develop the intricate optimization task into a practical method that iteratively applies the trained autoencoder and relatively simple computations that estimate and address the unknown degradation operator. We evaluate our method for blind inpainting where the goal is to recover training images from degradation of many missing pixels in an unknown pattern. We examine various deep autoencoder architectures, such as fully connected and U-Net (with various nonlinearities and at diverse train loss values), and show that our method significantly outperforms previous methods for training data recovery from autoencoders. Importantly, our method greatly improves the recovery performance also in settings that were previously considered highly challenging, and even impractical, for such retrieval.
Use Your INSTINCT: INSTruction optimization usIng Neural bandits Coupled with Transformers
Authors: Authors: Xiaoqiang Lin, Zhaoxuan Wu, Zhongxiang Dai, Wenyang Hu, Yao Shu, See-Kiong Ng, Patrick Jaillet, Bryan Kian Hsiang Low
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Abstract
Large language models (LLMs) have shown remarkable instruction-following capabilities and achieved impressive performances in various applications. However, the performances of LLMs depend heavily on the instructions given to them, which are typically manually tuned with substantial human efforts. Recent work has used the query-efficient Bayesian optimization (BO) algorithm to automatically optimize the instructions given to black-box LLMs. However, BO usually falls short when optimizing highly sophisticated (e.g., high-dimensional) objective functions, such as the functions mapping an instruction to the performance of an LLM. This is mainly due to the limited expressive power of the Gaussian process (GP) model which is used by BO as a surrogate to model the objective function. Meanwhile, it has been repeatedly shown that neural networks (NNs), especially pre-trained transformers, possess strong expressive power and can model highly complex functions. So, we adopt a neural bandit algorithm which replaces the GP in BO by an NN surrogate to optimize instructions for black-box LLMs. More importantly, the neural bandit algorithm allows us to naturally couple the NN surrogate with the hidden representation learned by a pre-trained transformer (i.e., an open-source LLM), which significantly boosts its performance. These motivate us to propose our INSTruction optimization usIng Neural bandits Coupled with Transformers} (INSTINCT) algorithm. We perform instruction optimization for ChatGPT and use extensive experiments to show that our INSTINCT consistently outperforms the existing methods in different tasks, such as in various instruction induction tasks and the task of improving the zero-shot chain-of-thought instruction.
Joint Network Lifetime Maximization and Relay Selection Design in Underwater Acoustic Sensor Networks
Authors: Authors: Z. Mohammadi, M. Soleimanpour-Moghadam, S. Talebi, H. Ahmadi
Subjects: Networking and Internet Architecture (cs.NI)
Abstract
The paper proposes a new approach to minimize the number of relays while maximizing the lifetime of underwater acoustic sensor networks (UASNs). This involves formulating the relay node placement (RNP) problem as a multi-objective optimization problem and employing the multi-objective lexico-graphic method (MOLM) to solve it. To achieve the optimal solution, the MOLM consists of two steps. First, the problem of lifetime maximization is tackled to find RNP solutions. This transforms the RNP into a non-convex optimization problem which is then converted into a convex programming equivalent. The proposed method has the same computational complexity as previous relay-node adjustment (RA) and difference convex algorithm (DCA) methods. The second step introduces a novel relay node selection to reach the optimal number of relays. Simulation results demonstrate that it has superior network lifetime and efficiency compared to RA and DCA.
Proximal Policy Optimization-Based Reinforcement Learning Approach for DC-DC Boost Converter Control: A Comparative Evaluation Against Traditional Control Techniques
Abstract
This article presents a proximal policy optimization (PPO) based reinforcement learning (RL) approach for DC-DC boost converter control, which is compared to traditional control methods. The performance of the PPO algorithm is evaluated using MATLAB Simulink co-simulation, and the results demonstrate that the most efficient approach for achieving short settling time and stability is to combine the PPO algorithm with reinforcement learning based control method. The simulation results indicate that the step response characteristics provided using the control method based on RL with the PPO algorithm outperform traditional control approaches, which can be used to improve DC-DC boost converter control. This research also highlights the inherent capability of the reinforcement learning method to enhance the performance of boost converter control.
Perceptual error optimization for Monte Carlo animation rendering
Authors: Authors: Miša Korać, Corentin Salaün, Iliyan Georgiev, Pascal Grittmann, Philipp Slusallek, Karol Myszkowski, Gurprit Singh
Abstract
Independently estimating pixel values in Monte Carlo rendering results in a perceptually sub-optimal white-noise distribution of error in image space. Recent works have shown that perceptual fidelity can be improved significantly by distributing pixel error as blue noise instead. Most such works have focused on static images, ignoring the temporal perceptual effects of animation display. We extend prior formulations to simultaneously consider the spatial and temporal domains, and perform an analysis to motivate a perceptually better spatio-temporal error distribution. We then propose a practical error optimization algorithm for spatio-temporal rendering and demonstrate its effectiveness in various configurations.
Co-Optimizing Cache Partitioning and Multi-Core Task Scheduling: Exploit Cache Sensitivity or Not?
Authors: Authors: Binqi Sun, Debayan Roy, Tomasz Kloda, Andrea Bastoni, Rodolfo Pellizzoni, Marco Caccamo
Subjects: Hardware Architecture (cs.AR); Distributed, Parallel, and Cluster Computing (cs.DC); Operating Systems (cs.OS)
Abstract
Cache partitioning techniques have been successfully adopted to mitigate interference among concurrently executing real-time tasks on multi-core processors. Considering that the execution time of a cache-sensitive task strongly depends on the cache available for it to use, co-optimizing cache partitioning and task allocation improves the system's schedulability. In this paper, we propose a hybrid multi-layer design space exploration technique to solve this multi-resource management problem. We explore the interplay between cache partitioning and schedulability by systematically interleaving three optimization layers, viz., (i) in the outer layer, we perform a breadth-first search combined with proactive pruning for cache partitioning; (ii) in the middle layer, we exploit a first-fit heuristic for allocating tasks to cores; and (iii) in the inner layer, we use the well-known recurrence relation for the schedulability analysis of non-preemptive fixed-priority (NP-FP) tasks in a uniprocessor setting. Although our focus is on NP-FP scheduling, we evaluate the flexibility of our framework in supporting different scheduling policies (NP-EDF, P-EDF) by plugging in appropriate analysis methods in the inner layer. Experiments show that, compared to the state-of-the-art techniques, the proposed framework can improve the real-time schedulability of NP-FP task sets by an average of 15.2% with a maximum improvement of 233.6% (when tasks are highly cache-sensitive) and a minimum of 1.6% (when cache sensitivity is low). For such task sets, we found that clustering similar-period (or mutually compatible) tasks often leads to higher schedulability (on average 7.6%) than clustering by cache sensitivity. In our evaluation, the framework also achieves good results for preemptive and dynamic-priority scheduling policies.
Dual Conic Proxies for AC Optimal Power Flow
Authors: Authors: Guancheng Qiu, Mathieu Tanneau, Pascal Van Hentenryck
Abstract
In recent years, there has been significant interest in the development of machine learning-based optimization proxies for AC Optimal Power Flow (AC-OPF). Although significant progress has been achieved in predicting high-quality primal solutions, no existing learning-based approach can provide valid dual bounds for AC-OPF. This paper addresses this gap by training optimization proxies for a convex relaxation of AC-OPF. Namely, the paper considers a second-order cone (SOC) relaxation of ACOPF, and proposes a novel dual architecture that embeds a fast, differentiable (dual) feasibility recovery, thus providing valid dual bounds. The paper combines this new architecture with a self-supervised learning scheme, which alleviates the need for costly training data generation. Extensive numerical experiments on medium- and large-scale power grids demonstrate the efficiency and scalability of the proposed methodology.
Variance Reduced Halpern Iteration for Finite-Sum Monotone Inclusions
Authors: Authors: Xufeng Cai, Ahmet Alacaoglu, Jelena Diakonikolas
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC)
Abstract
Machine learning approaches relying on such criteria as adversarial robustness or multi-agent settings have raised the need for solving game-theoretic equilibrium problems. Of particular relevance to these applications are methods targeting finite-sum structure, which generically arises in empirical variants of learning problems in these contexts. Further, methods with computable approximation errors are highly desirable, as they provide verifiable exit criteria. Motivated by these applications, we study finite-sum monotone inclusion problems, which model broad classes of equilibrium problems. Our main contributions are variants of the classical Halpern iteration that employ variance reduction to obtain improved complexity guarantees in which $n$ component operators in the finite sum are ``on average'' either cocoercive or Lipschitz continuous and monotone, with parameter $L$. The resulting oracle complexity of our methods, which provide guarantees for the last iterate and for a (computable) operator norm residual, is $\widetilde{\mathcal{O}}( n + \sqrt{n}L\varepsilon^{-1})$, which improves upon existing methods by a factor up to $\sqrt{n}$. This constitutes the first variance reduction-type result for general finite-sum monotone inclusions and for more specific problems such as convex-concave optimization when operator norm residual is the optimality measure. We further argue that, up to poly-logarithmic factors, this complexity is unimprovable in the monotone Lipschitz setting; i.e., the provided result is near-optimal.
IBCL: Zero-shot Model Generation for Task Trade-offs in Continual Learning
Authors: Authors: Pengyuan Lu, Michele Caprio, Eric Eaton, Insup Lee
Abstract
Like generic multi-task learning, continual learning has the nature of multi-objective optimization, and therefore faces a trade-off between the performance of different tasks. That is, to optimize for the current task distribution, it may need to compromise performance on some previous tasks. This means that there exist multiple models that are Pareto-optimal at different times, each addressing a distinct task performance trade-off. Researchers have discussed how to train particular models to address specific trade-off preferences. However, existing algorithms require training overheads proportional to the number of preferences -- a large burden when there are multiple, possibly infinitely many, preferences. As a response, we propose Imprecise Bayesian Continual Learning (IBCL). Upon a new task, IBCL (1) updates a knowledge base in the form of a convex hull of model parameter distributions and (2) obtains particular models to address task trade-off preferences with zero-shot. That is, IBCL does not require any additional training overhead to generate preference-addressing models from its knowledge base. We show that models obtained by IBCL have guarantees in identifying the Pareto optimal parameters. Moreover, experiments on standard image classification and NLP tasks support this guarantee. Statistically, IBCL improves average per-task accuracy by at most 23\% and peak per-task accuracy by at most 15\% with respect to the baseline methods, with steadily near-zero or positive backward transfer. Most importantly, IBCL significantly reduces the training overhead from training 1 model per preference to at most 3 models for all preferences.
Optimizing Key-Selection for Face-based One-Time Biometrics via Morphing
Authors: Authors: Daile Osorio-Roig, Mahdi Ghafourian, Christian Rathgeb, Ruben Vera-Rodriguez, Christoph Busch, Julian Fierrez
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Nowadays, facial recognition systems are still vulnerable to adversarial attacks. These attacks vary from simple perturbations of the input image to modifying the parameters of the recognition model to impersonate an authorised subject. So-called privacy-enhancing facial recognition systems have been mostly developed to provide protection of stored biometric reference data, i.e. templates. In the literature, privacy-enhancing facial recognition approaches have focused solely on conventional security threats at the template level, ignoring the growing concern related to adversarial attacks. Up to now, few works have provided mechanisms to protect face recognition against adversarial attacks while maintaining high security at the template level. In this paper, we propose different key selection strategies to improve the security of a competitive cancelable scheme operating at the signal level. Experimental results show that certain strategies based on signal-level key selection can lead to complete blocking of the adversarial attack based on an iterative optimization for the most secure threshold, while for the most practical threshold, the attack success chance can be decreased to approximately 5.0%.
Soft Convex Quantization: Revisiting Vector Quantization with Convex Optimization
Abstract
Vector Quantization (VQ) is a well-known technique in deep learning for extracting informative discrete latent representations. VQ-embedded models have shown impressive results in a range of applications including image and speech generation. VQ operates as a parametric K-means algorithm that quantizes inputs using a single codebook vector in the forward pass. While powerful, this technique faces practical challenges including codebook collapse, non-differentiability and lossy compression. To mitigate the aforementioned issues, we propose Soft Convex Quantization (SCQ) as a direct substitute for VQ. SCQ works like a differentiable convex optimization (DCO) layer: in the forward pass, we solve for the optimal convex combination of codebook vectors that quantize the inputs. In the backward pass, we leverage differentiability through the optimality conditions of the forward solution. We then introduce a scalable relaxation of the SCQ optimization and demonstrate its efficacy on the CIFAR-10, GTSRB and LSUN datasets. We train powerful SCQ autoencoder models that significantly outperform matched VQ-based architectures, observing an order of magnitude better image reconstruction and codebook usage with comparable quantization runtime.
Keyword: adam
AdaMerging: Adaptive Model Merging for Multi-Task Learning
Authors: Authors: Enneng Yang, Zhenyi Wang, Li Shen, Shiwei Liu, Guibing Guo, Xingwei Wang, Dacheng Tao
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Multi-task learning (MTL) aims to empower a model to tackle multiple tasks simultaneously. A recent development known as task arithmetic has revealed that several models, each fine-tuned for distinct tasks, can be directly merged into a single model to execute MTL without necessitating a retraining process using the initial training data. Nevertheless, this direct addition of models often leads to a significant deterioration in the overall performance of the merged model. This decline occurs due to potential conflicts and intricate correlations among the multiple tasks. Consequently, the challenge emerges of how to merge pre-trained models more effectively without using their original training data. This paper introduces an innovative technique called Adaptive Model Merging (AdaMerging). This approach aims to autonomously learn the coefficients for model merging, either in a task-wise or layer-wise manner, without relying on the original training data. Specifically, our AdaMerging method operates as an automatic, unsupervised task arithmetic scheme. It leverages entropy minimization on unlabeled test samples from the multi-task setup as a surrogate objective function to iteratively refine the merging coefficients of the multiple models. Our experimental findings across eight tasks demonstrate the efficacy of the AdaMerging scheme we put forth. Compared to the current state-of-the-art task arithmetic merging scheme, AdaMerging showcases a remarkable 11\% improvement in performance. Notably, AdaMerging also exhibits superior generalization capabilities when applied to unseen downstream tasks. Furthermore, it displays a significantly enhanced robustness to data distribution shifts that may occur during the testing phase.
Keyword: gradient
Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion
Abstract
Consistency Models (CM) (Song et al., 2023) accelerate score-based diffusion model sampling at the cost of sample quality but lack a natural way to trade-off quality for speed. To address this limitation, we propose Consistency Trajectory Model (CTM), a generalization encompassing CM and score-based models as special cases. CTM trains a single neural network that can -- in a single forward pass -- output scores (i.e., gradients of log-density) and enables unrestricted traversal between any initial and final time along the Probability Flow Ordinary Differential Equation (ODE) in a diffusion process. CTM enables the efficient combination of adversarial training and denoising score matching loss to enhance performance and achieves new state-of-the-art FIDs for single-step diffusion model sampling on CIFAR-10 (FID 1.73) and ImageNet at 64X64 resolution (FID 2.06). CTM also enables a new family of sampling schemes, both deterministic and stochastic, involving long jumps along the ODE solution trajectories. It consistently improves sample quality as computational budgets increase, avoiding the degradation seen in CM. Furthermore, CTM's access to the score accommodates all diffusion model inference techniques, including exact likelihood computation.
A Comparison of Mesh-Free Differentiable Programming and Data-Driven Strategies for Optimal Control under PDE Constraints
Authors: Authors: Roussel Desmond Nzoyem, David A.W. Barton, Tom Deakin
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
Abstract
The field of Optimal Control under Partial Differential Equations (PDE) constraints is rapidly changing under the influence of Deep Learning and the accompanying automatic differentiation libraries. Novel techniques like Physics-Informed Neural Networks (PINNs) and Differentiable Programming (DP) are to be contrasted with established numerical schemes like Direct-Adjoint Looping (DAL). We present a comprehensive comparison of DAL, PINN, and DP using a general-purpose mesh-free differentiable PDE solver based on Radial Basis Functions. Under Laplace and Navier-Stokes equations, we found DP to be extremely effective as it produces the most accurate gradients; thriving even when DAL fails and PINNs struggle. Additionally, we provide a detailed benchmark highlighting the limited conditions under which any of those methods can be efficiently used. Our work provides a guide to Optimal Control practitioners and connects them further to the Deep Learning community.
Computing a Sparse Approximate Inverse on Quantum Annealing Machines
Abstract
Many engineering problems involve solving large linear systems of equations. Conjugate gradient (CG) is one of the most popular iterative methods for solving such systems. However, CG typically requires a good preconditioner to speed up convergence. One such preconditioner is the sparse approximate inverse (SPAI). In this paper, we explore the computation of an SPAI on quantum annealing machines by solving a series of quadratic unconstrained binary optimization (QUBO) problems. Numerical experiments are conducted using both well-conditioned and poorly-conditioned linear systems arising from a 2D finite difference formulation of the Poisson problem.
Implicit regularization of multi-task learning and finetuning in overparameterized neural networks
Abstract
It is common in deep learning to train networks on auxiliary tasks with the expectation that the learning will transfer, at least partially, to another task of interest. In this work, we investigate the inductive biases that result from learning auxiliary tasks, either simultaneously (multi-task learning, MTL) or sequentially (pretraining and subsequent finetuning, PT+FT). In the simplified setting of two-layer diagonal linear networks trained with gradient descent, we identify implicit regularization penalties associated with MTL and PT+FT, both of which incentivize feature sharing between tasks and sparsity in learned task-specific features. Notably, our results imply that during finetuning, networks operate in a hybrid of the kernel (or "lazy") regime and the feature learning ("rich") regime identified in prior work. Moreover, PT+FT can exhibit a novel "nested feature learning" behavior not captured by either regime, which biases it to extract a sparse subset of the features learned during pretraining. In ReLU networks, we reproduce all of these qualitative behaviors. We also observe that PT+FT (but not MTL) is biased to learn features that are correlated with (but distinct from) those needed for the auxiliary task, while MTL is biased toward using identical features for both tasks. As a result, we find that in realistic settings, MTL generalizes better when comparatively little data is available for the task of interest, while PT+FT outperforms it with more data available. We show that our findings hold qualitatively for a deep architecture trained on image classification tasks. Our characterization of the nested feature learning regime also motivates a modification to PT+FT that we find empirically improves performance. Overall, our results shed light on the impact of auxiliary task learning and suggest ways to leverage it more effectively.
On the Parallel Complexity of Multilevel Monte Carlo in Stocahstic Gradient Descent
Abstract
In the stochastic gradient descent (SGD) for sequential simulations such as the neural stochastic differential equations, the Multilevel Monte Carlo (MLMC) method is known to offer better theoretical computational complexity compared to the naive Monte Carlo approach. However, in practice, MLMC scales poorly on massively parallel computing platforms such as modern GPUs, because of its large parallel complexity which is equivalent to that of the naive Monte Carlo method. To cope with this issue, we propose the delayed MLMC gradient estimator that drastically reduces the parallel complexity of MLMC by recycling previously computed gradient components from earlier steps of SGD. The proposed estimator provably reduces the average parallel complexity per iteration at the cost of a slightly worse per-iteration convergence rate. In our numerical experiments, we use an example of deep hedging to demonstrate the superior parallel complexity of our method compared to the standard MLMC in SGD.
OneAdapt: Fast Adaptation for Deep Learning Applications via Backpropagation
Abstract
Deep learning inference on streaming media data, such as object detection in video or LiDAR feeds and text extraction from audio waves, is now ubiquitous. To achieve high inference accuracy, these applications typically require significant network bandwidth to gather high-fidelity data and extensive GPU resources to run deep neural networks (DNNs). While the high demand for network bandwidth and GPU resources could be substantially reduced by optimally adapting the configuration knobs, such as video resolution and frame rate, current adaptation techniques fail to meet three requirements simultaneously: adapt configurations (i) with minimum extra GPU or bandwidth overhead; (ii) to reach near-optimal decisions based on how the data affects the final DNN's accuracy, and (iii) do so for a range of configuration knobs. This paper presents OneAdapt, which meets these requirements by leveraging a gradient-ascent strategy to adapt configuration knobs. The key idea is to embrace DNNs' differentiability to quickly estimate the accuracy's gradient to each configuration knob, called AccGrad. Specifically, OneAdapt estimates AccGrad by multiplying two gradients: InputGrad (i.e. how each configuration knob affects the input to the DNN) and DNNGrad (i.e. how the DNN input affects the DNN inference output). We evaluate OneAdapt across five types of configurations, four analytic tasks, and five types of input data. Compared to state-of-the-art adaptation schemes, OneAdapt cuts bandwidth usage and GPU usage by 15-59% while maintaining comparable accuracy or improves accuracy by 1-5% while using equal or fewer resources.
Feather: An Elegant Solution to Effective DNN Sparsification
Authors: Authors: Athanasios Glentis Georgoulakis, George Retsinas, Petros Maragos
Abstract
Neural Network pruning is an increasingly popular way for producing compact and efficient models, suitable for resource-limited environments, while preserving high performance. While the pruning can be performed using a multi-cycle training and fine-tuning process, the recent trend is to encompass the sparsification process during the standard course of training. To this end, we introduce Feather, an efficient sparse training module utilizing the powerful Straight-Through Estimator as its core, coupled with a new thresholding operator and a gradient scaling technique, enabling robust, out-of-the-box sparsification performance. Feather's effectiveness and adaptability is demonstrated using various architectures on the CIFAR dataset, while on ImageNet it achieves state-of-the-art Top-1 validation accuracy using the ResNet-50 architecture, surpassing existing methods, including more complex and computationally heavy ones, by a considerable margin. Code is publicly available at https://github.com/athglentis/feather .
Distributionally Safe Reinforcement Learning under Model Uncertainty: A Single-Level Approach by Differentiable Convex Programming
Authors: Authors: Alaa Eddine Chriat, Chuangchuang Sun
Subjects: Machine Learning (cs.LG); Robotics (cs.RO); Systems and Control (eess.SY)
Abstract
Safety assurance is uncompromisable for safety-critical environments with the presence of drastic model uncertainties (e.g., distributional shift), especially with humans in the loop. However, incorporating uncertainty in safe learning will naturally lead to a bi-level problem, where at the lower level the (worst-case) safety constraint is evaluated within the uncertainty ambiguity set. In this paper, we present a tractable distributionally safe reinforcement learning framework to enforce safety under a distributional shift measured by a Wasserstein metric. To improve the tractability, we first use duality theory to transform the lower-level optimization from infinite-dimensional probability space where distributional shift is measured, to a finite-dimensional parametric space. Moreover, by differentiable convex programming, the bi-level safe learning problem is further reduced to a single-level one with two sequential computationally efficient modules: a convex quadratic program to guarantee safety followed by a projected gradient ascent to simultaneously find the worst-case uncertainty. This end-to-end differentiable framework with safety constraints, to the best of our knowledge, is the first tractable single-level solution to address distributional safety. We test our approach on first and second-order systems with varying complexities and compare our results with the uncertainty-agnostic policies, where our approach demonstrates a significant improvement on safety guarantees.
SGORP: A Subgradient-based Method for d-Dimensional Rectilinear Partitioning
Authors: Authors: Muhammed Fatih Balin, Xiaojing An, Abdurrahman Yaşar, Ümit V. Çatalyürek
Abstract
Partitioning for load balancing is a crucial first step to parallelize any type of computation. In this work, we propose SGORP, a new spatial partitioning method based on Subgradient Optimization, to solve the $d$-dimensional Rectilinear Partitioning Problem (RPP). Our proposed method allows the use of customizable objective functions as well as some user-specific constraints, such as symmetric partitioning on selected dimensions. Extensive experimental evaluation using over 600 test matrices shows that our algorithm achieves favorable performance against the state-of-the-art RPP and Symmetric RPP algorithms. Additionally, we show the effectiveness of our algorithm to do application-specific load balancing using two applications as motivation: Triangle Counting and Sparse Matrix Multiplication (SpGEMM), where we model their load-balancing problems as $3$-dimensional RPPs.
Abstract
Conditional stochastic optimization has found applications in a wide range of machine learning tasks, such as invariant learning, AUPRC maximization, and meta-learning. As the demand for training models with large-scale distributed data grows in these applications, there is an increasing need for communication-efficient distributed optimization algorithms, such as federated learning algorithms. This paper considers the nonconvex conditional stochastic optimization in federated learning and proposes the first federated conditional stochastic optimization algorithm (FCSG) with a conditional stochastic gradient estimator and a momentum-based algorithm (FCSG-M). To match the lower bound complexity in the single-machine setting, we design an accelerated algorithm (Acc-FCSG-M) via the variance reduction to achieve the best sample and communication complexity. Compared with the existing optimization analysis for MAML in FL, federated conditional stochastic optimization considers the sample of tasks. Extensive experimental results on various tasks validate the efficiency of these algorithms.
Benign Overfitting and Grokking in ReLU Networks for XOR Cluster Data
Authors: Authors: Zhiwei Xu, Yutong Wang, Spencer Frei, Gal Vardi, Wei Hu
Abstract
Neural networks trained by gradient descent (GD) have exhibited a number of surprising generalization behaviors. First, they can achieve a perfect fit to noisy training data and still generalize near-optimally, showing that overfitting can sometimes be benign. Second, they can undergo a period of classical, harmful overfitting -- achieving a perfect fit to training data with near-random performance on test data -- before transitioning ("grokking") to near-optimal generalization later in training. In this work, we show that both of these phenomena provably occur in two-layer ReLU networks trained by GD on XOR cluster data where a constant fraction of the training labels are flipped. In this setting, we show that after the first step of GD, the network achieves 100% training accuracy, perfectly fitting the noisy labels in the training data, but achieves near-random test accuracy. At a later training step, the network achieves near-optimal test accuracy while still fitting the random labels in the training data, exhibiting a "grokking" phenomenon. This provides the first theoretical result of benign overfitting in neural network classification when the data distribution is not linearly separable. Our proofs rely on analyzing the feature learning process under GD, which reveals that the network implements a non-generalizable linear classifier after one step and gradually learns generalizable features in later steps.
Convergence Analysis and Latency Minimization for Semi-Federated Learning in Massive IoT Networks
Authors: Authors: Jianyang Ren, Wanli Ni, Hui Tian, Gaofeng Nie
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Abstract
As the number of sensors becomes massive in Internet of Things (IoT) networks, the amount of data is humongous. To process data in real-time while protecting user privacy, federated learning (FL) has been regarded as an enabling technique to push edge intelligence into IoT networks with massive devices. However, FL latency increases dramatically due to the increase of the number of parameters in deep neural network and the limited computation and communication capabilities of IoT devices. To address this issue, we propose a semi-federated learning (SemiFL) paradigm in which network pruning and over-the-air computation are efficiently applied. To be specific, each small base station collects the raw data from its served sensors and trains its local pruned model. After that, the global aggregation of local gradients is achieved through over-the-air computation. We first analyze the performance of the proposed SemiFL by deriving its convergence upper bound. To reduce latency, a convergence-constrained SemiFL latency minimization problem is formulated. By decoupling the original problem into several sub-problems, iterative algorithms are designed to solve them efficiently. Finally, numerical simulations are conducted to verify the effectiveness of our proposed scheme in reducing latency and guaranteeing the identification accuracy.
Semi-Federated Learning: Convergence Analysis and Optimization of A Hybrid Learning Framework
Authors: Authors: Jingheng Zheng, Wanli Ni, Hui Tian, Deniz Gunduz, Tony Q. S. Quek, Zhu Han
Subjects: Information Theory (cs.IT); Machine Learning (cs.LG); Signal Processing (eess.SP)
Abstract
Under the organization of the base station (BS), wireless federated learning (FL) enables collaborative model training among multiple devices. However, the BS is merely responsible for aggregating local updates during the training process, which incurs a waste of the computational resource at the BS. To tackle this issue, we propose a semi-federated learning (SemiFL) paradigm to leverage the computing capabilities of both the BS and devices for a hybrid implementation of centralized learning (CL) and FL. Specifically, each device sends both local gradients and data samples to the BS for training a shared global model. To improve communication efficiency over the same time-frequency resources, we integrate over-the-air computation for aggregation and non-orthogonal multiple access for transmission by designing a novel transceiver structure. To gain deep insights, we conduct convergence analysis by deriving a closed-form optimality gap for SemiFL and extend the result to two extra cases. In the first case, the BS uses all accumulated data samples to calculate the CL gradient, while a decreasing learning rate is adopted in the second case. Our analytical results capture the destructive effect of wireless communication and show that both FL and CL are special cases of SemiFL. Then, we formulate a non-convex problem to reduce the optimality gap by jointly optimizing the transmit power and receive beamformers. Accordingly, we propose a two-stage algorithm to solve this intractable problem, in which we provide the closed-form solutions to the beamformers. Extensive simulation results on two real-world datasets corroborate our theoretical analysis, and show that the proposed SemiFL outperforms conventional FL and achieves 3.2% accuracy gain on the MNIST dataset compared to state-of-the-art benchmarks.
ViT-ReciproCAM: Gradient and Attention-Free Visual Explanations for Vision Transformer
Authors: Authors: Seok-Yong Byun, Wonju Lee
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Abstract
This paper presents a novel approach to address the challenges of understanding the prediction process and debugging prediction errors in Vision Transformers (ViT), which have demonstrated superior performance in various computer vision tasks such as image classification and object detection. While several visual explainability techniques, such as CAM, Grad-CAM, Score-CAM, and Recipro-CAM, have been extensively researched for Convolutional Neural Networks (CNNs), limited research has been conducted on ViT. Current state-of-the-art solutions for ViT rely on class agnostic Attention-Rollout and Relevance techniques. In this work, we propose a new gradient-free visual explanation method for ViT, called ViT-ReciproCAM, which does not require attention matrix and gradient information. ViT-ReciproCAM utilizes token masking and generated new layer outputs from the target layer's input to exploit the correlation between activated tokens and network predictions for target classes. Our proposed method outperforms the state-of-the-art Relevance method in the Average Drop-Coherence-Complexity (ADCC) metric by $4.58\%$ to $5.80\%$ and generates more localized saliency maps. Our experiments demonstrate the effectiveness of ViT-ReciproCAM and showcase its potential for understanding and debugging ViT models. Our proposed method provides an efficient and easy-to-implement alternative for generating visual explanations, without requiring attention and gradient information, which can be beneficial for various applications in the field of computer vision.
On the Last-iterate Convergence in Time-varying Zero-sum Games: Extra Gradient Succeeds where Optimism Fails
Authors: Authors: Yi Feng, Hu Fu, Qun Hu, Ping Li, Ioannis Panageas, Bo Peng, Xiao Wang
Subjects: Computer Science and Game Theory (cs.GT)
Abstract
Last-iterate convergence has received extensive study in two player zero-sum games starting from bilinear, convex-concave up to settings that satisfy the MVI condition. Typical methods that exhibit last-iterate convergence for the aforementioned games include extra-gradient (EG) and optimistic gradient descent ascent (OGDA). However, all the established last-iterate convergence results hold for the restrictive setting where the underlying repeated game does not change over time. Recently, a line of research has focused on regret analysis of OGDA in time-varying games, i.e., games where payoffs evolve with time; the last-iterate behavior of OGDA and EG in time-varying environments remains unclear though. In this paper, we study the last-iterate behavior of various algorithms in two types of unconstrained, time-varying, bilinear zero-sum games: periodic and convergent perturbed games. These models expand upon the usual repeated game formulation and incorporate external environmental factors, such as the seasonal effects on species competition and vanishing external noise. In periodic games, we prove that EG will converge while OGDA and momentum method will diverge. This is quite surprising, as to the best of our knowledge, it is the first result that indicates EG and OGDA have qualitatively different last-iterate behaviors and do not exhibit similar behavior. In convergent perturbed games, we prove all these algorithms converge as long as the game itself stabilizes with a faster rate than $1/t$.
Tackling Hybrid Heterogeneity on Federated Optimization via Gradient Diversity Maximization
Abstract
Federated learning refers to a distributed machine learning paradigm in which data samples are decentralized and distributed among multiple clients. These samples may exhibit statistical heterogeneity, which refers to data distributions are not independent and identical across clients. Additionally, system heterogeneity, or variations in the computational power of the clients, introduces biases into federated learning. The combined effects of statistical and system heterogeneity can significantly reduce the efficiency of federated optimization. However, the impact of hybrid heterogeneity is not rigorously discussed. This paper explores how hybrid heterogeneity affects federated optimization by investigating server-side optimization. The theoretical results indicate that adaptively maximizing gradient diversity in server update direction can help mitigate the potential negative consequences of hybrid heterogeneity. To this end, we introduce a novel server-side gradient-based optimizer \textsc{FedAWARE} with theoretical guarantees provided. Intensive experiments in heterogeneous federated settings demonstrate that our proposed optimizer can significantly enhance the performance of federated learning across varying degrees of hybrid heterogeneity.
SHOT: Suppressing the Hessian along the Optimization Trajectory for Gradient-Based Meta-Learning
Abstract
In this paper, we hypothesize that gradient-based meta-learning (GBML) implicitly suppresses the Hessian along the optimization trajectory in the inner loop. Based on this hypothesis, we introduce an algorithm called SHOT (Suppressing the Hessian along the Optimization Trajectory) that minimizes the distance between the parameters of the target and reference models to suppress the Hessian in the inner loop. Despite dealing with high-order terms, SHOT does not increase the computational complexity of the baseline model much. It is agnostic to both the algorithm and architecture used in GBML, making it highly versatile and applicable to any GBML baseline. To validate the effectiveness of SHOT, we conduct empirical tests on standard few-shot learning tasks and qualitatively analyze its dynamics. We confirm our hypothesis empirically and demonstrate that SHOT outperforms the corresponding baseline. Code is available at: https://github.com/JunHoo-Lee/SHOT
Likelihood-Based Methods Improve Parameter Estimation in Opinion Dynamics Models
Authors: Authors: Jacopo Lenti, Gianmarco De Francisci Morales, Corrado Monti
Subjects: Social and Information Networks (cs.SI); Computers and Society (cs.CY)
Abstract
We show that a maximum likelihood approach for parameter estimation in agent-based models (ABMs) of opinion dynamics outperforms the typical simulation-based approach. Simulation-based approaches simulate the model repeatedly in search of a set of parameters that generates data similar enough to the observed one. In contrast, likelihood-based approaches derive a likelihood function that connects the unknown parameters to the observed data in a statistically principled way. We compare these two approaches on the well-known bounded-confidence model of opinion dynamics. We do so on three realistic scenarios of increasing complexity depending on data availability: (i) fully observed opinions and interactions, (ii) partially observed interactions, (iii) observed interactions with noisy proxies of the opinions. We highlight how identifying observed and latent variables is fundamental for connecting the model to the data. To realize the likelihood-based approach, we first cast the model into a probabilistic generative guise that supports a proper data likelihood. Then, we describe the three scenarios via probabilistic graphical models and show the nuances that go into translating the model. Finally, we implement the resulting probabilistic models in an automatic differentiation framework (PyTorch). This step enables easy and efficient maximum likelihood estimation via gradient descent. Our experimental results show that the maximum likelihood estimates are up to 4x more accurate and require up to 200x less computational time.
Discovering General Reinforcement Learning Algorithms with Adversarial Environment Design
Authors: Authors: Matthew Thomas Jackson, Minqi Jiang, Jack Parker-Holder, Risto Vuorio, Chris Lu, Gregory Farquhar, Shimon Whiteson, Jakob Nicolaus Foerster
Abstract
The past decade has seen vast progress in deep reinforcement learning (RL) on the back of algorithms manually designed by human researchers. Recently, it has been shown that it is possible to meta-learn update rules, with the hope of discovering algorithms that can perform well on a wide range of RL tasks. Despite impressive initial results from algorithms such as Learned Policy Gradient (LPG), there remains a generalization gap when these algorithms are applied to unseen environments. In this work, we examine how characteristics of the meta-training distribution impact the generalization performance of these algorithms. Motivated by this analysis and building on ideas from Unsupervised Environment Design (UED), we propose a novel approach for automatically generating curricula to maximize the regret of a meta-learned optimizer, in addition to a novel approximation of regret, which we name algorithmic regret (AR). The result is our method, General RL Optimizers Obtained Via Environment Design (GROOVE). In a series of experiments, we show that GROOVE achieves superior generalization to LPG, and evaluate AR against baseline metrics from UED, identifying it as a critical component of environment design in this setting. We believe this approach is a step towards the discovery of truly general RL algorithms, capable of solving a wide range of real-world environments.
Learning to Scale Logits for Temperature-Conditional GFlowNets
Authors: Authors: Minsu Kim, Joohwan Ko, Dinghuai Zhang, Ling Pan, Taeyoung Yun, Woochang Kim, Jinkyoo Park, Yoshua Bengio
Abstract
GFlowNets are probabilistic models that learn a stochastic policy that sequentially generates compositional structures, such as molecular graphs. They are trained with the objective of sampling such objects with probability proportional to the object's reward. Among GFlowNets, the temperature-conditional GFlowNets represent a family of policies indexed by temperature, and each is associated with the correspondingly tempered reward function. The major benefit of temperature-conditional GFlowNets is the controllability of GFlowNets' exploration and exploitation through adjusting temperature. We propose Learning to Scale Logits for temperature-conditional GFlowNets (LSL-GFN), a novel architectural design that greatly accelerates the training of temperature-conditional GFlowNets. It is based on the idea that previously proposed temperature-conditioning approaches introduced numerical challenges in the training of the deep network because different temperatures may give rise to very different gradient profiles and ideal scales of the policy's logits. We find that the challenge is greatly reduced if a learned function of the temperature is used to scale the policy's logits directly. We empirically show that our strategy dramatically improves the performances of GFlowNets, outperforming other baselines, including reinforcement learning and sampling methods, in terms of discovering diverse modes in multiple biochemical tasks.
Stable and Interpretable Deep Learning for Tabular Data: Introducing InterpreTabNet with the Novel InterpreStability Metric
Authors: Authors: Shiyun Wa, Xinai Lu, Minjuan Wang
Abstract
As Artificial Intelligence (AI) integrates deeper into diverse sectors, the quest for powerful models has intensified. While significant strides have been made in boosting model capabilities and their applicability across domains, a glaring challenge persists: many of these state-of-the-art models remain as black boxes. This opacity not only complicates the explanation of model decisions to end-users but also obstructs insights into intermediate processes for model designers. To address these challenges, we introduce InterpreTabNet, a model designed to enhance both classification accuracy and interpretability by leveraging the TabNet architecture with an improved attentive module. This design ensures robust gradient propagation and computational stability. Additionally, we present a novel evaluation metric, InterpreStability, which quantifies the stability of a model's interpretability. The proposed model and metric mark a significant stride forward in explainable models' research, setting a standard for transparency and interpretability in AI model design and application across diverse sectors. InterpreTabNet surpasses other leading solutions in tabular data analysis across varied application scenarios, paving the way for further research into creating deep-learning models that are both highly accurate and inherently explainable. The introduction of the InterpreStability metric ensures that the interpretability of future models can be measured and compared in a consistent and rigorous manner. Collectively, these contributions have the potential to promote the design principles and development of next-generation interpretable AI models, widening the adoption of interpretable AI solutions in critical decision-making environments.
CoLiDE: Concomitant Linear DAG Estimation
Authors: Authors: Seyed Saman Saboksayr, Gonzalo Mateos, Mariano Tepper
Abstract
We deal with the combinatorial problem of learning directed acyclic graph (DAG) structure from observational data adhering to a linear structural equation model (SEM). Leveraging advances in differentiable, nonconvex characterizations of acyclicity, recent efforts have advocated a continuous constrained optimization paradigm to efficiently explore the space of DAGs. Most existing methods employ lasso-type score functions to guide this search, which (i) require expensive penalty parameter retuning when the $\textit{unknown}$ SEM noise variances change across problem instances; and (ii) implicitly rely on limiting homoscedasticity assumptions. In this work, we propose a new convex score function for sparsity-aware learning of linear DAGs, which incorporates concomitant estimation of scale and thus effectively decouples the sparsity parameter from the exogenous noise levels. Regularization via a smooth, nonconvex acyclicity penalty term yields CoLiDE ($\textbf{Co}$ncomitant $\textbf{Li}$near $\textbf{D}$AG $\textbf{E}$stimation), a regression-based criterion amenable to efficient gradient computation and closed-form estimation of noise variances in heteroscedastic scenarios. Our algorithm outperforms state-of-the-art methods without incurring added complexity, especially when the DAGs are larger and the noise level profile is heterogeneous. We also find CoLiDE exhibits enhanced stability manifested via reduced standard deviations in several domain-specific metrics, underscoring the robustness of our novel linear DAG estimator.
ECoFLaP: Efficient Coarse-to-Fine Layer-Wise Pruning for Vision-Language Models
Abstract
Large Vision-Language Models (LVLMs) can understand the world comprehensively by integrating rich information from different modalities, achieving remarkable performance improvements on various multimodal downstream tasks. However, deploying LVLMs is often problematic due to their massive computational/energy costs and carbon consumption. Such issues make it infeasible to adopt conventional iterative global pruning, which is costly due to computing the Hessian matrix of the entire large model for sparsification. Alternatively, several studies have recently proposed layer-wise pruning approaches to avoid the expensive computation of global pruning and efficiently compress model weights according to their importance within a layer. However, these methods often suffer from suboptimal model compression due to their lack of a global perspective. To address this limitation in recent efficient pruning methods for large models, we propose Efficient Coarse-to-Fine Layer-Wise Pruning (ECoFLaP), a two-stage coarse-to-fine weight pruning approach for LVLMs. We first determine the sparsity ratios of different layers or blocks by leveraging the global importance score, which is efficiently computed based on the zeroth-order approximation of the global model gradients. Then, the multimodal model performs local layer-wise unstructured weight pruning based on globally-informed sparsity ratios. We validate our proposed method across various multimodal and unimodal models and datasets, demonstrating significant performance improvements over prevalent pruning techniques in the high-sparsity regime.
High-dimensional SGD aligns with emerging outlier eigenspaces
Abstract
We rigorously study the joint evolution of training dynamics via stochastic gradient descent (SGD) and the spectra of empirical Hessian and gradient matrices. We prove that in two canonical classification tasks for multi-class high-dimensional mixtures and either 1 or 2-layer neural networks, the SGD trajectory rapidly aligns with emerging low-rank outlier eigenspaces of the Hessian and gradient matrices. Moreover, in multi-layer settings this alignment occurs per layer, with the final layer's outlier eigenspace evolving over the course of training, and exhibiting rank deficiency when the SGD converges to sub-optimal classifiers. This establishes some of the rich predictions that have arisen from extensive numerical studies in the last decade about the spectra of Hessian and information matrices over the course of training in overparametrized networks.
Understanding In-Context Learning in Transformers and LLMs by Learning to Learn Discrete Functions
Authors: Authors: Satwik Bhattamishra, Arkil Patel, Phil Blunsom, Varun Kanade
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Abstract
In order to understand the in-context learning phenomenon, recent works have adopted a stylized experimental framework and demonstrated that Transformers can learn gradient-based learning algorithms for various classes of real-valued functions. However, the limitations of Transformers in implementing learning algorithms, and their ability to learn other forms of algorithms are not well understood. Additionally, the degree to which these capabilities are confined to attention-based models is unclear. Furthermore, it remains to be seen whether the insights derived from these stylized settings can be extrapolated to pretrained Large Language Models (LLMs). In this work, we take a step towards answering these questions by demonstrating the following: (a) On a test-bed with a variety of Boolean function classes, we find that Transformers can nearly match the optimal learning algorithm for 'simpler' tasks, while their performance deteriorates on more 'complex' tasks. Additionally, we find that certain attention-free models perform (almost) identically to Transformers on a range of tasks. (b) When provided a teaching sequence, i.e. a set of examples that uniquely identifies a function in a class, we show that Transformers learn more sample-efficiently. Interestingly, our results show that Transformers can learn to implement two distinct algorithms to solve a single task, and can adaptively select the more sample-efficient algorithm depending on the sequence of in-context examples. (c) Lastly, we show that extant LLMs, e.g. LLaMA-2, GPT-4, can compete with nearest-neighbor baselines on prediction tasks that are guaranteed to not be in their training set.
Keyword: super-resolution
Relaxed Octahedral Group Convolution for Learning Symmetry Breaking in 3D Physical Systems
Authors: Authors: Rui Wang, Robin Walters, Tess E.Smidt
Abstract
Deep equivariant models use symmetries to improve sample efficiency and generalization. However, the assumption of perfect symmetry in many of these models can sometimes be restrictive, especially when the data does not perfectly align with such symmetries. Thus, we introduce relaxed octahedral group convolution for modeling 3D physical systems in this paper. This flexible convolution technique provably allows the model to both maintain the highest level of equivariance that is consistent with data and discover the subtle symmetry-breaking factors in the physical systems. Empirical results validate that our approach can not only provide insights into the symmetry-breaking factors in phase transitions but also achieves superior performance in fluid super-resolution tasks.
Keyword: sgd
On the Parallel Complexity of Multilevel Monte Carlo in Stocahstic Gradient Descent
High-dimensional SGD aligns with emerging outlier eigenspaces
Keyword: optimization
Online Proactive Multi-Task Assignment with Resource Availability Anticipation
Reinforcement Learning from Automatic Feedback for High-Quality Unit Test Generation
Computing a Sparse Approximate Inverse on Quantum Annealing Machines
Learning Diverse Skills for Local Navigation under Multi-constraint Optimality
GenCO: Generating Diverse Solutions to Design Problems with Combinatorial Nature
Distributionally Safe Reinforcement Learning under Model Uncertainty: A Single-Level Approach by Differentiable Convex Programming
SGORP: A Subgradient-based Method for d-Dimensional Rectilinear Partitioning
A Recipe for Improved Certifiable Robustness: Capacity and Data
Parameterized Convex Minorant for Objective Function Approximation in Amortized Optimization
Federated Conditional Stochastic Optimization
Auto-FP: An Experimental Study of Automated Feature Preprocessing for Tabular Data
Tightly Joining Positioning and Control for Trustworthy Unmanned Aerial Vehicles Based on Factor Graph Optimization in Urban Transportation
A ModelOps-based Framework for Intelligent Medical Knowledge Extraction
Adaptive Spatio-Temporal Voxels Based Trajectory Planning for Autonomous Driving in Highway Traffic Flow
Long-Term Dynamic Window Approach for Kinodynamic Local Planning in Static and Crowd Environments
Exploring Federated Optimization by Reducing Variance of Adaptive Unbiased Client Sampling
Tackling Hybrid Heterogeneity on Federated Optimization via Gradient Diversity Maximization
Understanding Pan-Sharpening via Generalized Inverse
Efficient Quantification and Representation of Aggregate Flexibility in Electric Vehicles
Reward Model Ensembles Help Mitigate Overoptimization
SHOT: Suppressing the Hessian along the Optimization Trajectory for Gradient-Based Meta-Learning
R-LGP: A Reachability-guided Logic-geometric Programming Framework for Optimal Task and Motion Planning on Mobile Manipulators
Everest: GPU-Accelerated System For Mining Temporal Motifs
A Deep Instance Generative Framework for MILP Solvers Under Limited Data Availability
Incorporating Target Vehicle Trajectories Predicted by Deep Learning Into Model Predictive Controlled Vehicles
Magicremover: Tuning-free Text-guided Image inpainting with Diffusion Models
Approximating Robot Configuration Spaces with few Convex Sets using Clique Covers of Visibility Graphs
A Grammatical Compositional Model for Video Action Detection
CoLiDE: Concomitant Linear DAG Estimation
Recovery of Training Data from Overparameterized Autoencoders: An Inverse Problem Perspective
Use Your INSTINCT: INSTruction optimization usIng Neural bandits Coupled with Transformers
Joint Network Lifetime Maximization and Relay Selection Design in Underwater Acoustic Sensor Networks
Proximal Policy Optimization-Based Reinforcement Learning Approach for DC-DC Boost Converter Control: A Comparative Evaluation Against Traditional Control Techniques
Perceptual error optimization for Monte Carlo animation rendering
Co-Optimizing Cache Partitioning and Multi-Core Task Scheduling: Exploit Cache Sensitivity or Not?
Dual Conic Proxies for AC Optimal Power Flow
Variance Reduced Halpern Iteration for Finite-Sum Monotone Inclusions
IBCL: Zero-shot Model Generation for Task Trade-offs in Continual Learning
Optimizing Key-Selection for Face-based One-Time Biometrics via Morphing
Soft Convex Quantization: Revisiting Vector Quantization with Convex Optimization
Keyword: adam
AdaMerging: Adaptive Model Merging for Multi-Task Learning
Keyword: gradient
Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion
A Comparison of Mesh-Free Differentiable Programming and Data-Driven Strategies for Optimal Control under PDE Constraints
Computing a Sparse Approximate Inverse on Quantum Annealing Machines
Implicit regularization of multi-task learning and finetuning in overparameterized neural networks
On the Parallel Complexity of Multilevel Monte Carlo in Stocahstic Gradient Descent
OneAdapt: Fast Adaptation for Deep Learning Applications via Backpropagation
Feather: An Elegant Solution to Effective DNN Sparsification
Distributionally Safe Reinforcement Learning under Model Uncertainty: A Single-Level Approach by Differentiable Convex Programming
SGORP: A Subgradient-based Method for d-Dimensional Rectilinear Partitioning
Federated Conditional Stochastic Optimization
Benign Overfitting and Grokking in ReLU Networks for XOR Cluster Data
Convergence Analysis and Latency Minimization for Semi-Federated Learning in Massive IoT Networks
Semi-Federated Learning: Convergence Analysis and Optimization of A Hybrid Learning Framework
ViT-ReciproCAM: Gradient and Attention-Free Visual Explanations for Vision Transformer
On the Last-iterate Convergence in Time-varying Zero-sum Games: Extra Gradient Succeeds where Optimism Fails
Tackling Hybrid Heterogeneity on Federated Optimization via Gradient Diversity Maximization
SHOT: Suppressing the Hessian along the Optimization Trajectory for Gradient-Based Meta-Learning
Likelihood-Based Methods Improve Parameter Estimation in Opinion Dynamics Models
Discovering General Reinforcement Learning Algorithms with Adversarial Environment Design
Learning to Scale Logits for Temperature-Conditional GFlowNets
Stable and Interpretable Deep Learning for Tabular Data: Introducing InterpreTabNet with the Novel InterpreStability Metric
CoLiDE: Concomitant Linear DAG Estimation
ECoFLaP: Efficient Coarse-to-Fine Layer-Wise Pruning for Vision-Language Models
High-dimensional SGD aligns with emerging outlier eigenspaces
Understanding In-Context Learning in Transformers and LLMs by Learning to Learn Discrete Functions
Keyword: super-resolution
Relaxed Octahedral Group Convolution for Learning Symmetry Breaking in 3D Physical Systems