Burning RED: Unlocking Subtask-Driven Reinforcement Learning and Risk-Awareness in Average-Reward Markov Decision Processes

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Rojas, Juan Sebastian, Lee, Chi-Guhn
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911125934702592
author Rojas, Juan Sebastian
Lee, Chi-Guhn
author_facet Rojas, Juan Sebastian
Lee, Chi-Guhn
contents Average-reward Markov decision processes (MDPs) provide a foundational framework for sequential decision-making under uncertainty. However, average-reward MDPs have remained largely unexplored in reinforcement learning (RL) settings, with the majority of RL-based efforts having been allocated to discounted MDPs. In this work, we study a unique structural property of average-reward MDPs and utilize it to introduce Reward-Extended Differential (or RED) reinforcement learning: a novel RL framework that can be used to effectively and efficiently solve various learning objectives, or subtasks, simultaneously in the average-reward setting. We introduce a family of RED learning algorithms for prediction and control, including proven-convergent algorithms for the tabular case. We then showcase the power of these algorithms by demonstrating how they can be used to learn a policy that optimizes, for the first time, the well-known conditional value-at-risk (CVaR) risk measure in a fully-online manner, without the use of an explicit bi-level optimization scheme or an augmented state-space.
format Preprint
id arxiv_https___arxiv_org_abs_2410_10578
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Burning RED: Unlocking Subtask-Driven Reinforcement Learning and Risk-Awareness in Average-Reward Markov Decision Processes
Rojas, Juan Sebastian
Lee, Chi-Guhn
Machine Learning
Artificial Intelligence
Average-reward Markov decision processes (MDPs) provide a foundational framework for sequential decision-making under uncertainty. However, average-reward MDPs have remained largely unexplored in reinforcement learning (RL) settings, with the majority of RL-based efforts having been allocated to discounted MDPs. In this work, we study a unique structural property of average-reward MDPs and utilize it to introduce Reward-Extended Differential (or RED) reinforcement learning: a novel RL framework that can be used to effectively and efficiently solve various learning objectives, or subtasks, simultaneously in the average-reward setting. We introduce a family of RED learning algorithms for prediction and control, including proven-convergent algorithms for the tabular case. We then showcase the power of these algorithms by demonstrating how they can be used to learn a policy that optimizes, for the first time, the well-known conditional value-at-risk (CVaR) risk measure in a fully-online manner, without the use of an explicit bi-level optimization scheme or an augmented state-space.
title Burning RED: Unlocking Subtask-Driven Reinforcement Learning and Risk-Awareness in Average-Reward Markov Decision Processes
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2410.10578