Deep learning with intelligent opponent in fencing

• المؤلف الرئيسي: Loh, Qiao Yan
• مؤلفون آخرون: Seah Hock Soon
• التنسيق: Final Year Project
• اللغة: English
• منشور في: Nanyang Technological University 2022
• الموضوعات: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• الوصول للمادة أونلاين: https://hdl.handle.net/10356/156771

VR technologies have enabled the development of virtual sports games, which greatly enhance the accessibility of physical sport by making it available virtually. Considering there is currently no such game for fencing, this project aims to design and develop a VR fencing game. For now, we have designed and implemented a fencer avatar to act in the virtual environment and conducted several reinforcement learning experiments to train an intelligent fencing agent as the game opponent. Training of the agent is conducted in a self-play setting. The training result showed that there is a need to further improve the action space of the agent.