arXiv research

Hybrid RL algorithms improve offline and online RL in linear MDPs.

A simple approach to offline RL without additional complexity.

New method defends RL agents from poisoning attacks without MDP knowledge.

RL tackles decision making in unknown environments, focusing on efficiency and efficacy.

evo-RL combines evolutionary computation with reinforcement learning for better adaptability.

Survey of RL algorithms for dynamic environments.

Critic-regularized regression improves offline RL performance.

HTMRL uses HTM for RL, adapting faster to changing environments.

Study assesses whether RL algorithm personalizes treatment sequences.

Deep reinforcement learning (deep RL) has been successful in learning sophisticated behaviors automatically; however, the learning process requires a huge number of trials. In contrast, animals can learn new tasks in just a few trials, benefiting from their prior knowledge about the world. This paper seeks to bridge th…

Optimistic RL algorithms are simplified for deep RL with competitive performance.

This paper surveys RL for combinatorial optimization, focusing on TSP.

We introduce SLM Lab, a software framework for reproducible reinforcement learning (RL) research. SLM Lab implements a number of popular RL algorithms, provides synchronous and asynchronous parallel experiment execution, hyperparameter search, and result analysis. RL algorithms in SLM Lab are implemented in a modular w…

Reincarnating RL reuses prior work to accelerate RL progress.

This work bridges offline RL and DRL to address distributional shift.

RH-UCRL combines pessimism and optimism for robust RL.

New model-free DR-RL algorithm with finite sample complexity.

Many interesting real world domains involve reinforcement learning (RL) in partially observable environments. Efficient learning in such domains is important, but existing sample complexity bounds for partially observable RL are at least exponential in the episode length. We give, to our knowledge, the first partially …

PDCA algorithm learns policies for RL with constraints using a primal-dual approach.

Paper proposes a hybrid RL algorithm that combines offline and online data without needing reward info.

We propose a general formulation for addressing reinforcement learning (RL) problems in settings with observational data. That is, we consider the problem of learning good policies solely from historical data in which unobserved factors (confounders) affect both observed actions and rewards. Our formulation allows us t…

To overcome the curses of dimensionality and modeling of Dynamic Programming (DP) methods to solve Markov Decision Process (MDP) problems, Reinforcement Learning (RL) methods are adopted in practice. Contrary to traditional RL algorithms which do not consider the structural properties of the optimal policy, we propose …

FMDP-BF algorithm improves RL in factored MDPs with exponential regret reduction.

Improves RL algorithms with two techniques.

Deep neuroevolution and deep reinforcement learning (deep RL) algorithms are two popular approaches to policy search. The former is widely applicable and rather stable, but suffers from low sample efficiency. By contrast, the latter is more sample efficient, but the most sample efficient variants are also rather unstab…

New RL algorithm minimizes distributional learning error.

Survey of offline RL theory and practical algorithm design challenges.

This chapter reviews statistical tools for reinforcement learning.

Paper shows RLHF can be solved similarly to standard RL.

This paper bridges offline and online RL by studying policy finetuning with a reference policy.

New offline RL algorithms tackle partial data coverage with optimal performance and practicality.

Mirror descent method improved RL algorithms.

To overcome the curse of dimensionality and curse of modeling in Dynamic Programming (DP) methods for solving classical Markov Decision Process (MDP) problems, Reinforcement Learning (RL) algorithms are popular. In this paper, we consider an infinite-horizon average reward MDP problem and prove the optimality of the th…

RL models improve target control in SSGs for security applications.

Unified hybrid RL algorithm improves online RL performance with offline data.

PBCS combines RL and motion planning for better exploration.

Despite the recent progress in deep reinforcement learning field (RL), and, arguably because of it, a large body of work remains to be done in reproducing and carefully comparing different RL algorithms. We present catalyst.RL, an open source framework for RL research with a focus on reproducibility and flexibility. Ma…

SIBRE boosts reinforcement learning convergence by rewarding improvement over past performance.

We present a reduction from reinforcement learning (RL) to no-regret online learning based on the saddle-point formulation of RL, by which "any" online algorithm with sublinear regret can generate policies with provable performance guarantees. This new perspective decouples the RL problem into two parts: regret minimiz…

Reinforcement learning (RL) algorithms have made huge progress in recent years by leveraging the power of deep neural networks (DNN). Despite the success, deep RL algorithms are known to be sample inefficient, often requiring many rounds of interaction with the environments to obtain satisfactory performance. Recently,…

In this report, we present a new reinforcement learning (RL) benchmark based on the Sonic the Hedgehog (TM) video game franchise. This benchmark is intended to measure the performance of transfer learning and few-shot learning algorithms in the RL domain. We also present and evaluate some baseline algorithms on the new…

New RL algorithm ensures stable, replicable policies.

We aim to jointly optimize antenna tilt angle, and vertical and horizontal half-power beamwidths of the macrocells in a heterogeneous cellular network (HetNet). The interactions between the cells, most notably due to their coupled interference render this optimization prohibitively complex. Utilizing a single agent rei…

Off-policy reinforcement learning (RL) using a fixed offline dataset of logged interactions is an important consideration in real world applications. This paper studies offline RL using the DQN replay dataset comprising the entire replay experience of a DQN agent on 60 Atari 2600 games. We demonstrate that recent off-p…

Unified framework for finite-sample RL algorithms using Lyapunov theory.

New algorithm for offline RL with linear approx in MDPs and MGs, nearly optimal.

Distributional Reinforcement Learning (RL) differs from traditional RL in that, rather than the expectation of total returns, it estimates distributions and has achieved state-of-the-art performance on Atari Games. The key challenge in practical distributional RL algorithms lies in how to parameterize estimated distrib…

MOPO optimizes offline RL by penalizing dynamics uncertainty.