VIPO: Value Function Inconsistency Penalized Offline Reinforcement Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Xuyang, Yan, Keyu, Wang, Guojian, Zhao, Lin
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913120616710144
author Chen, Xuyang
Yan, Keyu
Wang, Guojian
Zhao, Lin
author_facet Chen, Xuyang
Yan, Keyu
Wang, Guojian
Zhao, Lin
contents Offline reinforcement learning (RL) learns effective policies from pre-collected datasets, offering a practical solution for applications where online interactions are risky or costly. Model-based approaches are particularly advantageous for offline RL, owing to their data efficiency and generalizability. However, due to inherent model errors, model-based methods often artificially introduce conservatism guided by heuristic uncertainty estimation, which can be unreliable. In this paper, we introduce VIPO, a novel model-based offline RL algorithm that incorporates self-supervised feedback from value estimation to enhance model training. Specifically, the model is learned by additionally minimizing the inconsistency between the value learned directly from the offline data and the value estimated from the model. We perform comprehensive evaluations from multiple perspectives to show that VIPO can learn a highly accurate model efficiently and consistently outperform existing methods. In particular, it achieves state-of-the-art performance on almost all tasks in both D4RL and NeoRL benchmarks. Overall, VIPO offers a general framework that can be readily integrated into existing model-based offline RL algorithms to systematically enhance model accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2504_11944
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle VIPO: Value Function Inconsistency Penalized Offline Reinforcement Learning
Chen, Xuyang
Yan, Keyu
Wang, Guojian
Zhao, Lin
Machine Learning
Artificial Intelligence
Offline reinforcement learning (RL) learns effective policies from pre-collected datasets, offering a practical solution for applications where online interactions are risky or costly. Model-based approaches are particularly advantageous for offline RL, owing to their data efficiency and generalizability. However, due to inherent model errors, model-based methods often artificially introduce conservatism guided by heuristic uncertainty estimation, which can be unreliable. In this paper, we introduce VIPO, a novel model-based offline RL algorithm that incorporates self-supervised feedback from value estimation to enhance model training. Specifically, the model is learned by additionally minimizing the inconsistency between the value learned directly from the offline data and the value estimated from the model. We perform comprehensive evaluations from multiple perspectives to show that VIPO can learn a highly accurate model efficiently and consistently outperform existing methods. In particular, it achieves state-of-the-art performance on almost all tasks in both D4RL and NeoRL benchmarks. Overall, VIPO offers a general framework that can be readily integrated into existing model-based offline RL algorithms to systematically enhance model accuracy.
title VIPO: Value Function Inconsistency Penalized Offline Reinforcement Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2504.11944