Search and Learning Algorithms on Othello

Implemented Monte-Carlo Tree Search, Greedy Search, and Convolutional Neural Network using keras and NumPy. Developed Othello agents that played against each other in all 100 consecutive games, with the search agent using alpha-beta pruning winning approximately 75% of the games. Analyzed efficiency of various search and learning algorithms by creating a round-robin tournament to simulate number of wins and losses for each agent.

Link to paper

Aknazar Janibek
software developer from almaty, kz
aknazar.janibek@gmail.com

github.com/ajanibekcode
linkedin.com/in/aknazar-janibek