Behavioral Priors and Dynamics Models: Improving Performance and Domain Transfer in Offline RL

by   Catherine Cang, et al.
berkeley college

Offline Reinforcement Learning (RL) aims to extract near-optimal policies from imperfect offline data without additional environment interactions. Extracting policies from diverse offline datasets has the potential to expand the range of applicability of RL by making the training process safer, faster, and more streamlined. We investigate how to improve the performance of offline RL algorithms, its robustness to the quality of offline data, as well as its generalization capabilities. To this end, we introduce Offline Model-based RL with Adaptive Behavioral Priors (MABE). Our algorithm is based on the finding that dynamics models, which support within-domain generalization, and behavioral priors, which support cross-domain generalization, are complementary. When combined together, they substantially improve the performance and generalization of offline RL policies. In the widely studied D4RL offline RL benchmark, we find that MABE achieves higher average performance compared to prior model-free and model-based algorithms. In experiments that require cross-domain generalization, we find that MABE outperforms prior methods. Our website is available at .


page 1

page 6


Domain Generalization for Robust Model-Based Offline Reinforcement Learning

Existing offline reinforcement learning (RL) algorithms typically assume...

Double Check Your State Before Trusting It: Confidence-Aware Bidirectional Offline Model-Based Imagination

The learned policy of model-free offline reinforcement learning (RL) met...

Offline Reinforcement Learning from Images with Latent Space Models

Offline reinforcement learning (RL) refers to the problem of learning po...

Offline Behaviors of Online Friends

In this work we analyze traces of mobility and co-location among a group...

Model-Based Offline Planning

Offline learning is a key part of making reinforcement learning (RL) use...

Offline RL Policies Should be Trained to be Adaptive

Offline RL algorithms must account for the fact that the dataset they ar...

Improving Offline RL by Blending Heuristics

We propose Heuristic Blending (HUBL), a simple performance-improving tec...

Please sign up or login with your details

Forgot password? Click here to reset