MaxInfoRL: Boosting exploration in reinforcement learning through information gain maximization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sukhija, Bhavya, Coros, Stelian, Krause, Andreas, Abbeel, Pieter, Sferrazza, Carmelo
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918108277506048
author Sukhija, Bhavya
Coros, Stelian
Krause, Andreas
Abbeel, Pieter
Sferrazza, Carmelo
author_facet Sukhija, Bhavya
Coros, Stelian
Krause, Andreas
Abbeel, Pieter
Sferrazza, Carmelo
contents Reinforcement learning (RL) algorithms aim to balance exploiting the current best strategy with exploring new options that could lead to higher rewards. Most common RL algorithms use undirected exploration, i.e., select random sequences of actions. Exploration can also be directed using intrinsic rewards, such as curiosity or model epistemic uncertainty. However, effectively balancing task and intrinsic rewards is challenging and often task-dependent. In this work, we introduce a framework, MaxInfoRL, for balancing intrinsic and extrinsic exploration. MaxInfoRL steers exploration towards informative transitions, by maximizing intrinsic rewards such as the information gain about the underlying task. When combined with Boltzmann exploration, this approach naturally trades off maximization of the value function with that of the entropy over states, rewards, and actions. We show that our approach achieves sublinear regret in the simplified setting of multi-armed bandits. We then apply this general formulation to a variety of off-policy model-free RL methods for continuous state-action spaces, yielding novel algorithms that achieve superior performance across hard exploration problems and complex scenarios such as visual control tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2412_12098
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MaxInfoRL: Boosting exploration in reinforcement learning through information gain maximization
Sukhija, Bhavya
Coros, Stelian
Krause, Andreas
Abbeel, Pieter
Sferrazza, Carmelo
Machine Learning
Artificial Intelligence
Robotics
Reinforcement learning (RL) algorithms aim to balance exploiting the current best strategy with exploring new options that could lead to higher rewards. Most common RL algorithms use undirected exploration, i.e., select random sequences of actions. Exploration can also be directed using intrinsic rewards, such as curiosity or model epistemic uncertainty. However, effectively balancing task and intrinsic rewards is challenging and often task-dependent. In this work, we introduce a framework, MaxInfoRL, for balancing intrinsic and extrinsic exploration. MaxInfoRL steers exploration towards informative transitions, by maximizing intrinsic rewards such as the information gain about the underlying task. When combined with Boltzmann exploration, this approach naturally trades off maximization of the value function with that of the entropy over states, rewards, and actions. We show that our approach achieves sublinear regret in the simplified setting of multi-armed bandits. We then apply this general formulation to a variety of off-policy model-free RL methods for continuous state-action spaces, yielding novel algorithms that achieve superior performance across hard exploration problems and complex scenarios such as visual control tasks.
title MaxInfoRL: Boosting exploration in reinforcement learning through information gain maximization
topic Machine Learning
Artificial Intelligence
Robotics
url https://arxiv.org/abs/2412.12098