Beyond Prioritized Replay: Sampling States in Model-Based RL via Simulated Priorities.
Jincheng MeiYangchen PanMartha WhiteAmir-massoud FarahmandHengshuai YaoPublished in: CoRR (2020)
Keyphrases
- model free
- reinforcement learning
- random sampling
- function approximation
- belief state
- multi agent
- state space
- optimal policy
- neural network
- markov decision processes
- learning algorithm
- state action
- initial state
- simulation model
- policy iteration
- possibilistic logic
- sampling methods
- long run
- state variables
- sample size
- dynamic programming
- learning process
- genetic algorithm