Game AI - IDG5301

Course Description

Game technology incorporates a number of core technical fields that are relevant for modern game development, design, and production. Most of these areas are driven by key artificial intelligence (AI) techniques such as expert domain-knowledge systems, search and optimization, and computational intelligence in games.

The primary goal of this unit is the understanding, design, implementation, and use of basic and nouvelle AI techniques for generating efficient intelligent behaviors in games. The study unit aims to introduce students to the theory of basic and advanced game artificial intelligence topics and provide hands-on experience on the implementation of popular algorithms on commercial-standard games.

Please, check regularly the course plan for detailed information on lectures, tutorials, and project plan. Note that the course plan is subject to changes.

Course Schedule

MSc in Digital Games 2022 | GameAI

Course Projects

During the course, students will have to complete two projects. The first project is not mandatory but recommended to complete as a softer introduction to game AI. The second project (see Final Assignment below) is mandatory. For the completion of the course, students must deliver two AI controllers and a written report.

Mid-Term Assignment: Ms. Pac-Man MCTS Agent

The Ms. Pac-man vs Ghost Team competition is a contest held at several congresses and conferences on machine learning around the world in which AI-controllers for Ms.Pac-man and the Ghost-team compete for the highest ranking. For this course, each student has to develop one MCTS Ms. Pac-man player to compete against the ghost team controllers included in the software package. This is a simulator entirely written in Java with a well-documented interface.

Each student must submit the code for a Ms. Pac-Man agent to Kosmas Pinitas (via e-mail: kosmas.pinitas@um.edu.mt ) by midnight on Friday, Nov 25. There will be a Ms. Pac-Man competition on Monday, Nov 28 where all generated agents will compete against each other to find out the weaknesses and strengths of the chosen designs and techniques.

Final Assignment: Two Custom AI Controllers & Written Report

Students must use two advanced AI methods (ANN+BP, GA, TDL, ANN+GA, TDL+ANN,...); to implement either two different agents or two levels of control for a single agent for one of the following competitions (MCTS cannot be used) :


A written report including the methodology used and an empirical evaluation of the two controllers has to be handed in along with a google drive or dropbox link containing the source code, a readme file with guidelines on how to run the 2 different approaches, and a video demonstrating the two controllers in action.

The written report must follow the given template (Word and LaTeX) and not exceed a maximum of 5 pages.

This assignment must be handed in by Jan 22 before 15:00 at the Institute of Digital Games.


Performance Assessment and Grading

The assessment of the course is based on three factors:

  • Final Project and Written Report (50%) - Students must complete an independent project and write a 5-page report on their progress and performance (see below).

  • Oral Examination (40%) - An oral exam will assess the knowledge of the students in the areas presented during the course, including Pathfinding (A* and beyond - Navigation Meshes, Artificial Potential Fields), Expert Domain-Knowledge Systems (Finite State Machines, Behavior Trees), Search Algorithms (Basic Search, Min-Max Search, Tree-Search, Monte-Carlo Tree Search), and Computational Intelligence and Games (Neural Networks, Genetic Algorithms, Temporal Difference Learning)

  • Presentation (10%) - As part of the oral exam, a short 10 minute presentation of their respective final project.