Importance-Weighted Variational Inference Model Estimation for Offline Bayesian Model-Based Reinforcement Learning

Importance-Weighted Variational Inference Model Estimation for Offline Bayesian Model-Based Reinforcement Learning

This paper proposes a model estimation method in offline Bayesian model-based reinforcement learning (MBRL). Learning a Bayes-adaptive Markov decision process (BAMDP) model using standard variational inference often suffers from poor predictive performance due to covariate shift between offline data...

Full description

Bibliographic Details
Main Authors: Toru Hishinuma, Kei Senda
Format: Article
Language:English
Published: IEEE 2023-01-01
Series:IEEE Access
Subjects:
Bayesian model-based reinforcement learning
decision-aware reinforcement learning
offline reinforcement learning
Online Access:https://ieeexplore.ieee.org/document/10368011/

Internet

https://ieeexplore.ieee.org/document/10368011/

Similar Items