Reinforcement Learning as Iterative and Amortised Inference.
Beren MillidgeAlexander TschantzAnil K. SethChristopher L. BuckleyPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- function approximation
- learning algorithm
- multi agent
- state space
- probabilistic inference
- model free
- bayesian inference
- inference engine
- robotic control
- temporal difference
- reinforcement learning methods
- iterative methods
- inference mechanism
- inference process
- bayesian model
- learning process
- bayesian networks
- policy search
- temporal difference learning
- transfer learning
- action selection
- supervised learning
- dynamic bayesian networks
- probabilistic reasoning
- graphical models
- optimal policy
- random fields