Reinforcement Learning for Creating Evaluation Function Using Convolutional Neural Network in Hex.
Kei TakadaHiroyuki IizukaMasahito YamamotoPublished in: TAAI (2017)
Keyphrases
- evaluation function
- convolutional neural network
- temporal difference
- game tree search
- reinforcement learning
- temporal difference learning
- face detection
- td learning
- state action
- reinforcement learning algorithms
- two player games
- function approximation
- game playing
- computer chess
- game tree
- alpha beta
- policy evaluation
- alpha beta pruning
- state space
- iterative deepening
- model free
- machine learning
- monte carlo tree search
- neural network
- optimality criterion
- detection method
- general game playing
- markov decision processes
- multi agent
- minimax search
- expected outcome
- function approximators
- action selection
- markov chain