Fiche de LE PELLETIER DE WOILLEMONT Pierre

Titre : Simulation Of Diverse Human Play-Styles In Video Games: A Reinforcement Learning Approach
Date : 11 septembre 2023
Heure: 16:00
Résumé : In this dissertation, we explore the application of reinforcement learning (RL) techniques to create human-like agents that can simulate diverse play-styles for use in automated game testing. We investigate how RL algorithms can be used to generate agents that mimic the behavior of human players and how these agents can be used to evaluate the game. Our objective is to bridge the gap between the needs and constraints of the video game industry and the capabilities and requirements of the existing machine learning algorithms in order to introduce methods realistically capable of creating relevant agents useful for automated tests of video games. To that effect we develloped three approaches which can be used at different stages of the video game production. We tested our approaches on a video game prototype develloped in Unity, which was also used to gather human data in order to measure the effectiveness of our approaches.
Lieu : Jussieu