Reinforcement Learning to Create Value and Policy Functions Using Minimax Tree Search in Hex.
Kei TakadaHiroyuki IizukaMasahito YamamotoPublished in: IEEE Trans. Games (2020)
Keyphrases
- tree search
- game tree search
- alpha beta
- game tree
- reinforcement learning
- optimal policy
- state space
- board game
- branch and bound
- constraint propagation
- search tree
- mathematical programming
- search algorithm
- action space
- evaluation function
- iterative deepening
- action selection
- imperfect information
- function approximation
- game playing
- reinforcement learning algorithms
- depth first search
- temporal difference
- monte carlo
- machine learning
- function approximators
- search space
- dynamic programming
- heuristic search
- search strategies
- objective function
- reinforcement learning methods