Automated gameplay testing using reinforcement learning

In this research project, we would like to improve game testing using Reinforcement Learning (RL). By using RL, we can create intelligent agents that can play games independently and give us valuable feedback on the game’s quality. This will make game testing more effective and efficient. We also plan to explore a new approach called offline RL, where we can reuse data from previous game builds instead of having to interact with the game environment directly. This research project will enhance the game testing approach, which leads to better-quality games and a more streamlined development process.

Faculty Supervisor:

David Meger

Student:

Partner:

Ubisoft Divertissement

Discipline:

Computer science

Sector:

Information and cultural industries; Manufacturing

University:

McGill University

Program: