An Empirical Evaluation of Posterior Sampling for Constrained Reinforcement Learning.
Danil ProvodinPratik GajaneMykola PechenizkiyMaurits KapteinPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- markov chain monte carlo
- random sampling
- function approximation
- metropolis hastings
- reinforcement learning algorithms
- posterior probability
- dynamic programming
- learning algorithm
- state space
- probability distribution
- sampling strategies
- multi agent reinforcement learning
- bayesian framework
- temporal difference
- posterior distribution
- machine learning
- temporal difference learning
- sampling methods
- model free
- gaussian process
- neural network
- markov decision processes
- monte carlo
- parameter estimation
- importance sampling
- transfer learning
- sampling strategy
- sample size
- policy search
- markov chain