GDC is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

View, browse and sort the ever-growing list of sessions by pass type, track, and format. With this Session Viewer, you can view GDC 2023 session details, speakers and share your favorites via social media. You will be able to build your schedule and access it during the show via export or with the Mobile App, once live. Sessions do fill up and seating is first come, first serve, so arrive early to sessions that you would like to attend.

AI Summit: The Dimensional Curse of AAA Game Balancing: Data-Efficient and Robust Reinforcement Learning

Edgar Handy  (Machine Learning Engineer, SQUARE ENIX CO., LTD.)

Pass Type: All Access Pass, Summits Pass

Topic: Programming

Format: Lecture

Vault Recording: TBD

Audience Level: Intermediate

Game balancing is an important aspect of quality assurance in game development, yet it's time and manpower expensive. In particular, there have been increasing practical and theoretical examples of using reinforcement learning to optimize game mechanics balancing.

However, current state-of-the-art reinforcement learning algorithms still face some key challenges, which render them less reliable. First, it requires huge volume of data and expensive hardware to learn complex game mechanics. Second, game simulators are unstable while under development process. Third, rapidly changing game mechanics that renders previously trained model useless.

In this session, Edgar Handy, Machine Learning Engineer at Square Enix, introduces a set of solutions to address these key challenges as a step to reliable ML framework for game balancing, applied to their latest flagship AAA title battle mechanics, resulting in a cooperation between human developers and AI on improving battle mechanics.

Takeaway

Attendees will gain an understanding about the current key challenges in implementing reinforcement learning for balancing AAA game battle mechanics, and the method to address them.

Intended Audience

The is for AI/gameplay programmers, game designers, and those who are generally interested in game balancing/QA for AAA games. Basic knowledge of programming and reinforcement learning are not pre-requisite, but would be helpful to have.