Temporal difference learning of N-tuple networks for the game 2048.
Marcin Grzegorz SzubertWojciech JaskowskiPublished in: CIG (2014)
Keyphrases
- temporal difference learning
- game playing
- function approximation
- monte carlo tree search
- fixed point
- approximate value iteration
- reinforcement learning
- video games
- evaluation function
- game play
- temporal difference
- educational games
- markov decision process
- monte carlo
- reinforcement learning algorithms
- optimal policy
- linear combination
- dynamic programming