Since reinforcement learning algorithms are notoriously data-intensive, the task of sampling
observations from the environment is usually split across multiple agents. However …

We present new policy mirror descent (PMD) methods for solving reinforcement learning
(RL) problems with either strongly convex or general convex regularizers. By exploring the …

We consider infinite-horizon discounted Markov decision processes and study the
convergence rates of the natural policy gradient (NPG) and the Q-NPG methods with the log …

Modern policy optimization methods in reinforcement learning, such as TRPO and PPO, owe
their success to the use of parameterized policies. However, while theoretical guarantees …

Markov Decision Processes are classically solved using Value Iteration and Policy Iteration
algorithms. Recent interest in Reinforcement Learning has motivated the study of methods …

This paper develops an unified framework to study finite-sample convergence guarantees of
a large class of value-based asynchronous reinforcement learning (RL) algorithms. We do …

This paper develops a unified Lyapunov framework for finite-sample analysis of a Markovian
stochastic approximation (SA) algorithm under a contraction operator with respect to an …

We introduce algorithms based on natural actor-critic and analyze their sample complexity
for solving two player zero-sum Markov games in the tabular case. Our results improve the …

Actor–critic style two-time-scale algorithms are one of the most popular methods in
reinforcement learning, and have seen great empirical success. However, their performance …

In this work, we study policy-based methods for solving the reinforcement learning problem,
where off-policy sampling and linear function approximation are employed for policy …