Skip to content
This a repository for pacman agents created as part of my MSc Data Science programme's modules in artificial intelligence and machine learning.
Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
QLearningAgents.py
README
decisionTree.py
mdpAgents.py
searchAlgorithmAgents.py

README

This repository contains four files, each of which consists of a single coursework submission for my MSc in Data Science covering modules in Artificial Intelligence and Machine Learning.

Each file contains python code for an agent which can play Pacman within the Berkeley AI Pacman game.
http://ai.berkeley.edu/home.html

To use these files, you will need to download the Pacman game from the above URL, save these files in the same directory and launch the game.

searchAlgorithmAgents.py
- This file contains an agent who uses a pathfinding algorithm (BFS) to find food whilst avoiding ghosts that come within a defined perimeter of the agenbt

mdpAgents.py
- This file contains an agent who uses a markov decision process to select an action

decisionTree.py
- This file contains an implentation from scratch of a decision tree with pruning. The decision tree is used to classify moves within the game based on data from previous games.

QLearningAgents.py
- This file contains an implementation of a pacman that agent that uses reinforcement learning (q learning)
You can’t perform that action at this time.