Rethinking Model-based, Policy-based, and Value-based Reinforcement Learning via the Lens of Representation Complexity.
Guhao FengHan ZhongPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- optimal policy
- model free
- decision problems
- state space
- action selection
- markov decision process
- reinforcement learning algorithms
- multi agent
- worst case
- action space
- markov decision processes
- multiscale
- function approximation
- computational complexity
- temporal difference
- partially observable
- policy gradient
- policy search
- state and action spaces