Search Results for author:
Found 49 papers, 3 papers with code
Global Optimality without Mixing Time Oracles in Average-reward RL via Multi-level Actor-Critic
In the context of average-reward reinforcement learning, the requirement for oracle knowledge of the mixing time, a measure of the duration a Markov chain under a fixed policy needs to achieve its stationary distribution-poses a significant challenge for the global convergence of policy gradient methods.
Independent RL for Cooperative-Competitive Agents: A Mean-Field Perspective
This MFTG NE is then shown to be $\mathcal{O}(1/M)$-NE for the finite population game where $M$ is a lower bound on the number of agents in each team.
Beyond Joint Demonstrations: Personalized Expert Guidance for Efficient Multi-Agent Reinforcement Learning
Multi-Agent Reinforcement Learning (MARL) algorithms face the challenge of efficient exploration due to the exponential increase in the size of the joint state-action space.
MaxMin-RLHF: Towards Equitable Alignment of Large Language Models with Diverse Human Preferences
Reinforcement Learning from Human Feedback (RLHF) aligns language models to human preferences by employing a singular reward model derived from preference data.
Learning Payment-Free Resource Allocation Mechanisms
We consider the design of mechanisms that allocate limited resources among self-interested agents using neural networks.
Byzantine-Resilient Decentralized Multi-Armed Bandits
In decentralized cooperative multi-armed bandits (MAB), each agent observes a distinct stream of rewards, and seeks to exchange information with others to select a sequence of arms so as to minimize its regret.
PARL: A Unified Framework for Policy Alignment in Reinforcement Learning
We present a novel unified bilevel optimization-based framework, \textsf{PARL}, formulated to address the recently highlighted critical issue of policy alignment in reinforcement learning using utility or preference-based feedback.
Limited-Memory Greedy Quasi-Newton Method with Non-asymptotic Superlinear Convergence Rate
Interestingly, our established non-asymptotic superlinear convergence rate demonstrates an explicit trade-off between the convergence speed and memory requirement, which to our knowledge, is the first of its kind.
Ada-NAV: Adaptive Trajectory Length-Based Sample Efficient Policy Learning for Robotic Navigation
Trajectory length stands as a crucial hyperparameter within reinforcement learning (RL) algorithms, significantly contributing to the sample inefficiency in robotics applications.
A Gradient-based Approach for Online Robust Deep Neural Network Training with Noisy Labels
In this paper, we propose a novel gradient-based approach to enable the detection of noisy labels for the online learning of model parameters, named Online Gradient-based Robust Selection (OGRS).
Sharpened Lazy Incremental Quasi-Newton Method
Specifically, they exhibit a superlinear rate with $O(d^2)$ cost in contrast to the linear rate of first-order methods with $O(d)$ cost and the quadratic rate of second-order methods with $O(d^3)$ cost.
Scalable Multi-Agent Reinforcement Learning with General Utilities
The objective is to find a localized policy that maximizes the average of the team's local utility functions without the full observability of each agent in the team.
Multi-agent Reinforcement Learning
reinforcement-learning
+1
Beyond Exponentially Fast Mixing in Average-Reward Reinforcement Learning via Multi-Level Monte Carlo Actor-Critic
Many existing reinforcement learning (RL) methods employ stochastic gradient iteration on the back end, whose stability hinges upon a hypothesis that the data-generating process mixes exponentially fast with a rate parameter that appears in the step-size selection.
STEERING: Stein Information Directed Exploration for Model-Based Reinforcement Learning
Directed Exploration is a crucial challenge in reinforcement learning (RL), especially when rewards are sparse.
Model-based Reinforcement Learning
reinforcement-learning
+1
Oracle-free Reinforcement Learning in Mean-Field Games along a Single Sample Path
Given that the underlying Markov Decision Process (MDP) of the agent is communicating, we provide finite sample convergence guarantees in terms of convergence of the mean-field and control policy to the mean-field equilibrium.
FedBC: Calibrating Global and Local Models via Federated Learning Beyond Consensus
In this work, we quantitatively calibrate the performance of global and local models in federated learning through a multi-criterion optimization-based framework, which we cast as a constrained program.
Dealing with Sparse Rewards in Continuous Control Robotics via Heavy-Tailed Policies
In this paper, we present a novel Heavy-Tailed Stochastic Policy Gradient (HT-PSG) algorithm to deal with the challenges of sparse rewards in continuous control problems.
Posterior Coreset Construction with Kernelized Stein Discrepancy for Model-Based Reinforcement Learning
Model-based approaches to reinforcement learning (MBRL) exhibit favorable performance in practice, but their theoretical guarantees in large spaces are mostly restricted to the setting when transition model is Gaussian or Lipschitz, and demands a posterior estimate whose representational complexity grows unbounded with time.
On the Hidden Biases of Policy Mirror Ascent in Continuous Action Spaces
Doing so incurs a persistent bias that appears in the attenuation rate of the expected policy gradient norm, which is inversely proportional to the radius of the action space.
Occupancy Information Ratio: Infinite-Horizon, Information-Directed, Parameterized Policy Search
In this work, we propose an information-directed objective for infinite-horizon reinforcement learning (RL), called the occupancy information ratio (OIR), inspired by the information ratio objectives used in previous information-directed sampling schemes for multi-armed bandits and Markov decision processes as well as recent advances in general utility RL.
Online, Informative MCMC Thinning with Kernelized Stein Discrepancy
A fundamental challenge in Bayesian inference is efficient representation of a target distribution.
Convergence Rates of Average-Reward Multi-agent Reinforcement Learning via Randomized Linear Programming
We establish that the sample complexity to obtain near-globally optimal solutions matches tight dependencies on the cardinality of the state and action spaces, and exhibits classical scalings with respect to the network in accordance with multi-agent optimization.
Multi-agent Reinforcement Learning
Reinforcement Learning (RL)
Achieving Zero Constraint Violation for Constrained Reinforcement Learning via Primal-Dual Approach
To achieve that, we advocate the use of randomized primal-dual approach to solve the CMDP problems and propose a conservative stochastic primal-dual algorithm (CSPDA) which is shown to exhibit $\tilde{\mathcal{O}}\left(1/\epsilon^2\right)$ sample complexity to achieve $\epsilon$-optimal cumulative reward with zero constraint violations.
Wasserstein-Splitting Gaussian Process Regression for Heterogeneous Online Bayesian Inference
Gaussian processes (GPs) are a well-known nonparametric Bayesian inference technique, but they suffer from scalability problems for large sample sizes, and their performance can degrade for non-stationary or spatially heterogeneous data.
On the Sample Complexity and Metastability of Heavy-tailed Policy Search in Continuous Control
To close this gap, we step towards persistent exploration in continuous space through policy parameterizations defined by distributions of heavier tails defined by tail-index parameter alpha, which increases the likelihood of jumping in state space.
MARL with General Utilities via Decentralized Shadow Reward Actor-Critic
DSAC augments the classic critic step by requiring agents to (i) estimate their local occupancy measure in order to (ii) estimate the derivative of the local utility with respect to their occupancy measure, i. e., the "shadow reward".
Composable Learning with Sparse Kernel Representations
We present a reinforcement learning algorithm for learning sparse non-parametric controllers in a Reproducing Kernel Hilbert Space.
Sparse Representations of Positive Functions via First and Second-Order Pseudo-Mirror Descent
We consider expected risk minimization problems when the range of the estimator is required to be nonnegative, motivated by the settings of maximum likelihood estimation (MLE) and trajectory optimization.
A Markov Decision Process Approach to Active Meta Learning
In supervised learning, we fit a single statistical model to a given data set, assuming that the data is associated with a singular task, which yields well-tuned models for specific use, but does not adapt well to new contexts.
Variational Policy Gradient Method for Reinforcement Learning with General Utilities
Analogously to the Policy Gradient Theorem \cite{sutton2000policy} available for RL with cumulative rewards, we derive a new Variational Policy Gradient Theorem for RL with general utilities, which establishes that the parametrized policy gradient may be obtained as the solution of a stochastic saddle point problem involving the Fenchel dual of the utility function.
Balancing Rates and Variance via Adaptive Batch-Size for Stochastic Optimization Problems
Stochastic gradient descent is a canonical tool for addressing stochastic optimization problems, and forms the bedrock of modern machine learning and statistics.
Efficient Large-Scale Gaussian Process Bandits by Believing only Informative Actions
Experimentally, we observe state of the art accuracy and complexity tradeoffs for GP bandit algorithms on various hyper-parameter tuning tasks, suggesting the merits of managing the complexity of GPs in bandit settings
Consistent Online Gaussian Process Regression Without the Sample Complexity Bottleneck
Gaussian processes provide a framework for nonlinear nonparametric Bayesian inference widely applicable across science and engineering.
Policy Gradient using Weak Derivatives for Reinforcement Learning
This paper considers policy search in continuous state-action reinforcement learning problems.
Collaborative Beamforming Under Localization Errors: A Discrete Optimization Approach
In particular, we formulate a discrete optimization problem to choose only a subset of agents to transmit the message signal so that the variance of the signal-to-noise ratio (SNR) received by the base station is minimized while the expected SNR exceeds a desired threshold.
Regret and Belief Complexity Trade-off in Gaussian Process Bandits via Information Thresholding
Doing so permits us to precisely characterize the trade-off between regret bounds of GP bandit algorithms and complexity of the posterior distributions depending on the compression parameter $\epsilon$ for both discrete and continuous action sets.
Cautious Reinforcement Learning via Distributional Risk in the Dual Domain
To ameliorate this issue, we propose a new definition of risk, which we call caution, as a penalty function added to the dual objective of the linear programming (LP) formulation of reinforcement learning.
On the Sample Complexity of Actor-Critic Method for Reinforcement Learning with Function Approximation
Actor-critic algorithms combine the merits of both approaches by alternating between steps to estimate the value function and policy gradient updates.
Optimally Compressed Nonparametric Online Learning
Batch training of machine learning models based on neural networks is now well established, whereas to date streaming methods are largely based on linear models.
Nonstationary Nonparametric Online Learning: Balancing Dynamic Regret and Model Parsimony
Prior works control dynamic regret growth only for linear models.
Adaptive Kernel Learning in Heterogeneous Networks
We consider learning in decentralized heterogeneous networks: agents seek to minimize a convex functional that aggregates data across the network, while only having access to their local data streams.
Global Convergence of Policy Gradient Methods to (Almost) Locally Optimal Policies
Under a further strict saddle points assumption, this result establishes convergence to essentially locally-optimal policies of the underlying problem, and thus bridges the gap in existing literature on the convergence of PG methods.
Nonparametric Stochastic Compositional Gradient Descent for Q-Learning in Continuous Markov Decision Problems
We consider Markov Decision Problems defined over continuous state and action spaces, where an autonomous agent seeks to learn a map from its states to actions so as to maximize its long-term discounted accumulation of rewards.
Decentralized Online Learning with Kernels
That is, we establish that with constant step-size selections agents' functions converge to a neighborhood of the globally optimal one while satisfying the consensus constraints as the penalty parameter is increased.
Parsimonious Online Learning with Kernels via Sparse Projections in Function Space
Despite their attractiveness, popular perception is that techniques for nonparametric function approximation do not scale to streaming data due to an intractable growth in the amount of storage they require.
Proximity Without Consensus in Online Multi-Agent Optimization
To do so, we depart from the canonical decentralized optimization framework where agreement constraints are enforced, and instead formulate a problem where each agent minimizes a global objective while enforcing network proximity constraints.
Multiagent Systems Systems and Control Computation
A Class of Parallel Doubly Stochastic Algorithms for Large-Scale Learning
Algorithms that are parallel in either of these dimensions exist, but RAPSA is the first attempt at a methodology that is parallel in both the selection of blocks and the selection of elements of the training set.
Decentralized Dynamic Discriminative Dictionary Learning
We consider discriminative dictionary learning in a distributed online setting, where a network of agents aims to learn a common set of dictionary elements of a feature space and model parameters while sequentially receiving observations.
Doubly Random Parallel Stochastic Methods for Large Scale Learning
Algorithms that are parallel in either of these dimensions exist, but RAPSA is the first attempt at a methodology that is parallel in both, the selection of blocks and the selection of elements of the training set.
