Recurrent Off-Policy Deep Reinforcement Learning Doesn't Have to be Slow

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Clark, Tyler, Evers, Christine, Hare, Jonathon
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914217640067072
author Clark, Tyler
Evers, Christine
Hare, Jonathon
author_facet Clark, Tyler
Evers, Christine
Hare, Jonathon
contents Recurrent off-policy deep reinforcement learning models achieve state-of-the-art performance but are often sidelined due to their high computational demands. In response, we introduce RISE (Recurrent Integration via Simplified Encodings), a novel approach that can leverage recurrent networks in any image-based off-policy RL setting without significant computational overheads via using both learnable and non-learnable encoder layers. When integrating RISE into leading non-recurrent off-policy RL algorithms, we observe a 35.6% human-normalized interquartile mean (IQM) performance improvement across the Atari benchmark. We analyze various implementation strategies to highlight the versatility and potential of our proposed framework.
format Preprint
id arxiv_https___arxiv_org_abs_2512_20513
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Recurrent Off-Policy Deep Reinforcement Learning Doesn't Have to be Slow
Clark, Tyler
Evers, Christine
Hare, Jonathon
Machine Learning
Recurrent off-policy deep reinforcement learning models achieve state-of-the-art performance but are often sidelined due to their high computational demands. In response, we introduce RISE (Recurrent Integration via Simplified Encodings), a novel approach that can leverage recurrent networks in any image-based off-policy RL setting without significant computational overheads via using both learnable and non-learnable encoder layers. When integrating RISE into leading non-recurrent off-policy RL algorithms, we observe a 35.6% human-normalized interquartile mean (IQM) performance improvement across the Atari benchmark. We analyze various implementation strategies to highlight the versatility and potential of our proposed framework.
title Recurrent Off-Policy Deep Reinforcement Learning Doesn't Have to be Slow
topic Machine Learning
url https://arxiv.org/abs/2512.20513