Federated Offline Reinforcement Learning: Collaborative Single-Policy Coverage Suffices.
Jiin WooLaixi ShiGauri JoshiYuejie ChiPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- collaborative learning
- markov decision process
- multi user
- learning process
- partially observable
- knowledge sharing
- control policy
- single user
- control policies
- action selection
- function approximation
- markov decision processes
- multi agent
- markov decision problems
- partially observable environments
- machine learning
- function approximators
- action space
- computer supported collaborative learning
- dynamic programming
- search engine