Saved in:
Bibliographic Details
Main Authors: Lyu, Guwen, Sato, Masahiro
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.15525
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909435852488704
author Lyu, Guwen
Sato, Masahiro
author_facet Lyu, Guwen
Sato, Masahiro
contents Back-stepping experience replay (BER) is a reinforcement learning technique that can accelerate learning efficiency in reversible environments. BER trains an agent with generated back-stepping transitions of collected experiences and normal forward transitions. However, the original algorithm is designed for a dense-reward environment that does not require complex exploration, limiting the BER technique to demonstrate its full potential. Herein, we propose an enhanced version of BER called Generalized BER (GBER), which extends the original algorithm to sparse-reward environments, particularly those with complex structures that require the agent to explore. GBER improves the performance of BER by introducing relabeling mechanism and applying diverse sampling strategies. We evaluate our modified version, which is based on a goal-conditioned deep deterministic policy gradient offline learning algorithm, across various maze navigation environments. The experimental results indicate that the GBER algorithm can significantly boost the performance and stability of the baseline algorithm in various sparse-reward environments, especially those with highly structural symmetricity.
format Preprint
id arxiv_https___arxiv_org_abs_2412_15525
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Generalized Back-Stepping Experience Replay in Sparse-Reward Environments
Lyu, Guwen
Sato, Masahiro
Machine Learning
Artificial Intelligence
Back-stepping experience replay (BER) is a reinforcement learning technique that can accelerate learning efficiency in reversible environments. BER trains an agent with generated back-stepping transitions of collected experiences and normal forward transitions. However, the original algorithm is designed for a dense-reward environment that does not require complex exploration, limiting the BER technique to demonstrate its full potential. Herein, we propose an enhanced version of BER called Generalized BER (GBER), which extends the original algorithm to sparse-reward environments, particularly those with complex structures that require the agent to explore. GBER improves the performance of BER by introducing relabeling mechanism and applying diverse sampling strategies. We evaluate our modified version, which is based on a goal-conditioned deep deterministic policy gradient offline learning algorithm, across various maze navigation environments. The experimental results indicate that the GBER algorithm can significantly boost the performance and stability of the baseline algorithm in various sparse-reward environments, especially those with highly structural symmetricity.
title Generalized Back-Stepping Experience Replay in Sparse-Reward Environments
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2412.15525