Saved in:
Bibliographic Details
Main Authors: Crowder, Douglas C., Trappett, Matthew L., McKenzie, Darrien M., Chance, Frances S.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.24016
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929570334113792
author Crowder, Douglas C.
Trappett, Matthew L.
McKenzie, Darrien M.
Chance, Frances S.
author_facet Crowder, Douglas C.
Trappett, Matthew L.
McKenzie, Darrien M.
Chance, Frances S.
contents Hindsight experience replay (HER) is well-known to accelerate goal-based reinforcement learning (RL). While HER is generally applied to off-policy RL algorithms, we previously showed that HER can also accelerate on-policy algorithms, such as proximal policy optimization (PPO), for goal-based Predator-Prey environments. Here, we show that we can improve the previous PPO-HER algorithm by selectively applying HER in a principled manner.
format Preprint
id arxiv_https___arxiv_org_abs_2410_24016
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Maximum Entropy Hindsight Experience Replay
Crowder, Douglas C.
Trappett, Matthew L.
McKenzie, Darrien M.
Chance, Frances S.
Machine Learning
Hindsight experience replay (HER) is well-known to accelerate goal-based reinforcement learning (RL). While HER is generally applied to off-policy RL algorithms, we previously showed that HER can also accelerate on-policy algorithms, such as proximal policy optimization (PPO), for goal-based Predator-Prey environments. Here, we show that we can improve the previous PPO-HER algorithm by selectively applying HER in a principled manner.
title Maximum Entropy Hindsight Experience Replay
topic Machine Learning
url https://arxiv.org/abs/2410.24016