README.md

Pac-Man with Markov Decision Process Solver

This repo acts as a documentation of my coursework as part of Artificial Intelligence Reasoning and Decision Making, which requires students to write code to control Pacman and win games using an MDP-solver.
Explore the docs »

View Demo · Report Bug · Request Feature

pacman-mdp

The Value-iteration Algorithm is the essential engine to support Pac-Man assessing situations and making most benetficial decision at each step.

Algorithm Development

Without Learning

With Learned Reward Function

A simple environment

A complete environment

With a comprehensive Learning and Rewarding Mechanism