Near-Minimax-Optimal Distributional Reinforcement Learning with a Generative Model.
Mark RowlandLi Kevin WenliangRémi MunosClare LyleYunhao TangWill DabneyPublished in: CoRR (2024)
Keyphrases
- generative model
- reinforcement learning
- probabilistic model
- prior knowledge
- bayesian framework
- worst case
- dynamic programming
- semi supervised
- posterior probability
- em algorithm
- learning algorithm
- latent dirichlet allocation
- image processing
- discriminative learning
- expectation maximization
- state space
- topic models
- fisher kernel
- pitman yor process