Deep Reinforcement Learning in Continuous Action Spaces: a Case Study in the Game of Simulated Curling.
Kyowoon LeeSol-A. KimJaesik ChoiSeong-Whan LeePublished in: ICML (2018)
Keyphrases
- action space
- reinforcement learning
- state space
- markov decision processes
- state and action spaces
- continuous state
- continuous state spaces
- real valued
- action selection
- reinforcement learning methods
- continuous action
- control policies
- markov decision process
- function approximation
- reinforcement learning algorithms
- stochastic processes
- stochastic games
- single agent
- reinforcement learning problems
- dynamic programming
- real robot
- state variables
- function approximators
- nash equilibrium
- learning algorithm
- temporal difference
- skill learning
- heuristic search
- markov decision problems
- state action
- multi agent
- infinite horizon
- partially observable