Whenever a clinician reflects on the efficacy of a sequence of treatment decisions for a
patient, they may try to identify critical time steps where, had they made different decisions …

We study episodic reinforcement learning in non-stationary linear (aka low-rank) Markov
Decision Processes (MDPs), ie, both the reward and transition kernel are linear with respect …

In most practical applications of reinforcement learning, it is untenable to maintain direct
estimates for individual states; in continuous-state systems, it is impossible. Instead …

Optimistic initialization underpins many theoretically sound exploration schemes in tabular
domains; however, in the deep function approximation setting, optimism can quickly …

A core operation in reinforcement learning (RL) is finding an action that is optimal with
respect to a learned value function. This operation is often challenging when the learned …

This paper evaluates adaptive Q-learning (AQL) and single-partition adaptive Q-learning
(SPAQL), two algorithms for efficient model-free episodic reinforcement learning (RL), in two …

Abstract Solving Partially Observable Markov Decision Processes (POMDPs) with
continuous actions is challenging, particularly for high-dimensional action spaces. To …

Solving continuous Partially Observable Markov Decision Processes (POMDPs) is
challenging, particularly for high-dimensional continuous action spaces. To alleviate this …

Reinforcement learning has received significant interest in recent years, due primarily to the
successes of deep reinforcement learning at solving many challenging tasks such as …

We study reinforcement learning in continuous state and action spaces endowed with a
metric. We provide a refined analysis of the algorithm of Sinclair, Banerjee, and Yu (2019) …