Offline Reinforcement Learning
Offline reinforcement learning (also known as Batch RL) studies how to learn optimal policies from pre-collected static datasets without any additional interaction with the environment. This is a key technology for bringing RL to high-stakes real-world applications such as healthcare, autonomous driving, and robotics.
Contents:
- Offline RL Overview — Distribution shift, CQL, IQL, TD3+BC, Decision Transformer, Offline-to-Online fine-tuning