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 day, pass type, topic, and format. With this Session Viewer, you can view GDC 2023 session details and 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 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.

LIGHTSPEED STUDIOS Developer Summit: Multi-agent AI Production of FPS Game Based on Reinforcement Learning (Presented by LIGHTSPEED STUDIOS)

Gerry SUN  (Senior Game AI Researcher, LIGHTSPEED STUDIOS, LIGHTSPEED STUDIOS)

Zhichao Shu  (Senior Game AI Researcher, LIGHTSPEED STUDIOS, LIGHTSPEED STUDIOS)

Location: Room 3022, West Hall

Date: Tuesday, March 21

Time: 3:00 pm - 3:30 pm

Pass Type: All Access Pass, Core Pass, Summits Pass, Expo Pass, Audio Pass, Independent Games Summit Pass

Topic: Programming

Format: Sponsor Developer Summit

Vault Recording: Video

Audience Level: Intermediate

LIGHTSPEED STUDIOS

In-game AI is a vital element in the creation of FPS games, as it can be implemented in various situations including PVE gameplay, to take over for disconnected players, and populate online matchmaking services. These circumstances necessitate a high level of both AI power and human-like behavior, thus creating a challenge for classical rule-based AI solutions.

Gerry and Zhichao will be delivering a presentation to discuss the implementation of a reinforcement learning AI solution that LIGHTSPEED STUDIOS is incorporating into an FPS game. This solution requires minimal human replay to create training data and has been demonstrated to be more efficient than imitation learning AI solution. The reinforcement learning AI proves to have a win rate that surpasses 90% when pitted against behavioral tree AI, and its performance throughout the game has been highly acclaimed by game developers.

Takeaway

Attendees will be informed of the advantages of the reinforcement learning AI solution in comparison to the traditional rule-based AI solutions, as well as its applications within the realm of game development.

Intended Audience

This talk is geared towards AI developers searching for effective solutions to create in-game AI and any individuals who have an interest in the topic of in-game AI.