/PML

Final Project from the course "Probabilistic Machine Learning" @ Data Science & Scientific Computing, University of Trieste, year 2020/2021, written in ipynb.

Primary LanguageJupyter Notebook

Reinforcement Learning for TicTacToe using Q-Learning

Final Project from the course "Probabilistic Machine Learning" @ Data Science & Scientific Computing, University of Trieste, year 2020/2021, written in ipynb.

General Info

This project implements an unbeatable TicTacToe X or O player. It was trained by using Bellmann's equation for a Markov Decision Process. The pdf report explains the theory behind the implementation.