Monte-Carlo Tree Search in Continuous Action Spaces with Value Gradients.
Jongmin LeeWonseok JeonGeon-Hyeong KimKee-Eung KimPublished in: AAAI (2020)
Keyphrases
- action space
- monte carlo tree search
- reinforcement learning methods
- state space
- markov decision processes
- real valued
- monte carlo
- reinforcement learning
- continuous action
- continuous state
- stochastic processes
- action selection
- evaluation function
- temporal difference
- single agent
- function approximators
- game tree
- temporal difference learning
- optimal policy
- markov chain
- markov decision problems
- dynamic programming