Board game and Supervised learning

[StepHaze]

Hello!
can I use your wonderful deep learning neural network library for my task:
there's a board game (with sowing mechanism: Mancala, Oware, etc.) and I have thousands of games of top players (with moves and result). I would like to train a game neural network using Supervised learning.
How is it possible? Please point me in the right direction.

[joaopauloschuler]

Hello @StepHaze ,
Applying NN for playing games is certainly possible. There are currently 234 papers at https://paperswithcode.com/task/atari-games about this subject.

I do not have any source code example with CAI solving this problem.

[StepHaze]

Thank you for your answer.
I'm not sure that Atari games are what I need to implement a supervised learning for a game of perfect information.

[joaopauloschuler]

I would have a look at https://paperswithcode.com/task/game-of-chess

[StepHaze]

Thanks. It's all about reinforcement learning. I would like to use a supervised learning.