Belief-Based Offline Reinforcement Learning for Delay-Robust Policy Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhan, Simon Sinong, Wu, Qingyuan, Wang, Philip, Yang, Frank, Shi, Xiangyu, Huang, Chao, Zhu, Qi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908827053457408
author Zhan, Simon Sinong
Wu, Qingyuan
Wang, Philip
Yang, Frank
Shi, Xiangyu
Huang, Chao
Zhu, Qi
author_facet Zhan, Simon Sinong
Wu, Qingyuan
Wang, Philip
Yang, Frank
Shi, Xiangyu
Huang, Chao
Zhu, Qi
contents Offline-to-online deployment of reinforcement-learning (RL) agents must bridge two gaps: (1) the sim-to-real gap, where real systems add latency and other imperfections not present in simulation, and (2) the interaction gap, where policies trained purely offline face out-of-distribution states during online execution because gathering new interaction data is costly or risky. Agents therefore have to generalize from static, delay-free datasets to dynamic, delay-prone environments. Standard offline RL learns from delay-free logs yet must act under delays that break the Markov assumption and hurt performance. We introduce DT-CORL (Delay-Transformer belief policy Constrained Offline RL), an offline-RL framework built to cope with delayed dynamics at deployment. DT-CORL (i) produces delay-robust actions with a transformer-based belief predictor even though it never sees delayed observations during training, and (ii) is markedly more sample-efficient than naïve history-augmentation baselines. Experiments on D4RL benchmarks with several delay settings show that DT-CORL consistently outperforms both history-augmentation and vanilla belief-based methods, narrowing the sim-to-real latency gap while preserving data efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2506_00131
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Belief-Based Offline Reinforcement Learning for Delay-Robust Policy Optimization
Zhan, Simon Sinong
Wu, Qingyuan
Wang, Philip
Yang, Frank
Shi, Xiangyu
Huang, Chao
Zhu, Qi
Machine Learning
Artificial Intelligence
Offline-to-online deployment of reinforcement-learning (RL) agents must bridge two gaps: (1) the sim-to-real gap, where real systems add latency and other imperfections not present in simulation, and (2) the interaction gap, where policies trained purely offline face out-of-distribution states during online execution because gathering new interaction data is costly or risky. Agents therefore have to generalize from static, delay-free datasets to dynamic, delay-prone environments. Standard offline RL learns from delay-free logs yet must act under delays that break the Markov assumption and hurt performance. We introduce DT-CORL (Delay-Transformer belief policy Constrained Offline RL), an offline-RL framework built to cope with delayed dynamics at deployment. DT-CORL (i) produces delay-robust actions with a transformer-based belief predictor even though it never sees delayed observations during training, and (ii) is markedly more sample-efficient than naïve history-augmentation baselines. Experiments on D4RL benchmarks with several delay settings show that DT-CORL consistently outperforms both history-augmentation and vanilla belief-based methods, narrowing the sim-to-real latency gap while preserving data efficiency.
title Belief-Based Offline Reinforcement Learning for Delay-Robust Policy Optimization
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.00131