DefaultVR: the AI Expansion. An application of artificial intelligence in competitive gaming and virtual reality.
Year: 2023
Type: dissertation
Abstract: This works presents the development of DefaultVR: The AI Expansion, an \nexpansion of the first degree thesis DefaultVR, a virtual reality tactical shooter \nonline game. In particular, this expansion aims to include an artificial agent, \ncalled Eve, capable of learning to play in the virtual reality world through \nreinforcement learning techniques. The agent learns to navigate the game \nenvironment, make decisions based on pseudo-visual information, and optimize its actions to maximize rewards. The development utilizes a deep \nreinforcement learning framework with the Proximal Policy Optimization \nalgorithm included with Units’s ML-Agents. \nExtensive experiments were conducted to evaluate the agent’s performance, comparing it against itself and human players. The results demonstrate the agent’s ability to adapt and improve over time, achieving competitive gameplay skills comparable to both new and experienced human \nVR players. The training process involved iterative optimization and analysis of various hyperparameters, observations’ and actions’ spaces, and \ntraining configurations. \nThe successful development of the artificial agent has significant implications for the field of gaming AI, showcasing its potential for creating engaging and challenging gameplay experiences. The research contributes to \nthe broader understanding of reinforcement learning techniques and their \napplication in training intelligent agents for real-world tasks.
Author: Levita, Luca
Institution: University of Bologna
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7065165975
