Successor-Predecessor Intrinsic Exploration

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yu, Changmin, Burgess, Neil, Sahani, Maneesh, Gershman, Samuel J.
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909082512785408
author Yu, Changmin
Burgess, Neil
Sahani, Maneesh
Gershman, Samuel J.
author_facet Yu, Changmin
Burgess, Neil
Sahani, Maneesh
Gershman, Samuel J.
contents Exploration is essential in reinforcement learning, particularly in environments where external rewards are sparse. Here we focus on exploration with intrinsic rewards, where the agent transiently augments the external rewards with self-generated intrinsic rewards. Although the study of intrinsic rewards has a long history, existing methods focus on composing the intrinsic reward based on measures of future prospects of states, ignoring the information contained in the retrospective structure of transition sequences. Here we argue that the agent can utilise retrospective information to generate explorative behaviour with structure-awareness, facilitating efficient exploration based on global instead of local information. We propose Successor-Predecessor Intrinsic Exploration (SPIE), an exploration algorithm based on a novel intrinsic reward combining prospective and retrospective information. We show that SPIE yields more efficient and ethologically plausible exploratory behaviour in environments with sparse rewards and bottleneck states than competing methods. We also implement SPIE in deep reinforcement learning agents, and show that the resulting agent achieves stronger empirical performance than existing methods on sparse-reward Atari games.
format Preprint
id arxiv_https___arxiv_org_abs_2305_15277
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Successor-Predecessor Intrinsic Exploration
Yu, Changmin
Burgess, Neil
Sahani, Maneesh
Gershman, Samuel J.
Machine Learning
Exploration is essential in reinforcement learning, particularly in environments where external rewards are sparse. Here we focus on exploration with intrinsic rewards, where the agent transiently augments the external rewards with self-generated intrinsic rewards. Although the study of intrinsic rewards has a long history, existing methods focus on composing the intrinsic reward based on measures of future prospects of states, ignoring the information contained in the retrospective structure of transition sequences. Here we argue that the agent can utilise retrospective information to generate explorative behaviour with structure-awareness, facilitating efficient exploration based on global instead of local information. We propose Successor-Predecessor Intrinsic Exploration (SPIE), an exploration algorithm based on a novel intrinsic reward combining prospective and retrospective information. We show that SPIE yields more efficient and ethologically plausible exploratory behaviour in environments with sparse rewards and bottleneck states than competing methods. We also implement SPIE in deep reinforcement learning agents, and show that the resulting agent achieves stronger empirical performance than existing methods on sparse-reward Atari games.
title Successor-Predecessor Intrinsic Exploration
topic Machine Learning
url https://arxiv.org/abs/2305.15277