Uniform State Abstraction For Reinforcement Learning.
John BurdenDaniel KudenkoPublished in: CoRR (2020)
Keyphrases
- state abstraction
- reinforcement learning
- state space
- markov decision processes
- hierarchical reinforcement learning
- reinforcement learning agents
- path finding
- reinforcement learning algorithms
- function approximation
- model free
- initial state
- multi agent
- learning algorithm
- optimal policy
- heuristic search
- transfer learning
- sufficient conditions
- finite state
- temporal difference
- partially observable
- policy iteration
- dynamic programming