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, topic, and format. With this Session Scheduler, you can build your schedule in advance and access it during the show via export or with the GDC Mobile App, once live. Sessions do fill up so please arrive early to sessions that you would like to attend. Adding a session to your schedule does not guarantee you a seat.

Applying Reinforcement Learning to Develop Game AI in NetEase Games

Tangjie Lyu (Senior Engineer, NetEase)

Pass Type: All Access Pass, Core+Summits Pass, Core Pass - Get your pass now!

Topic: Programming

Format: Session

Vault Recording: TBD

Audience Level: All

This session introduces the application of reinforcement learning in NetEase Games, including the problems encountered in the development, the tried solutions and the final results. It not only gives some advice, but also provides several tools, a series of solutions and a set of development process specifications for game developers to overcome difficulties when using this kind of technology. In NetEase Games, reinforcement learning demonstrates its ability to allow game designers to develop more intelligent, human-like AI. The results in real online games show that it has surpassed the original behavioral tree AI in some aspects and won unanimous praise from the project team and game players.


The attendee can get some advice of using reinforcement learning to develop game AI, including how to reduce training difficulty, improve training efficiency and how to make more interesting AI

Intended Audience

The audience who are interested in game AI and reinforcement learning