Stable Reinforcement Learning with Unbounded State Space.
Devavrat ShahQiaomin XieZhi XuPublished in: CoRR (2020)
Keyphrases
- state space
- reinforcement learning
- reinforcement learning algorithms
- markov decision processes
- heuristic search
- function approximation
- optimal policy
- state abstraction
- action space
- partially observable
- dynamic programming
- continuous state spaces
- dynamical systems
- markov decision process
- markov chain
- state variables
- planning problems
- state transition
- temporal difference
- multi agent
- neural network
- learning problems
- optimal control
- supervised learning
- search space
- control problems
- policy search
- reward function
- particle filter
- learning algorithm
- complex domains
- model free
- learning agent
- finite state
- function approximators
- reinforcement learning methods
- pattern databases
- continuous state
- learning process