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.

Use the GDC 2021 Session Viewer to browse and sort the ever-growing list of sessions by day, time, pass type, topic, and format. All registered attendees will be able to build their personal schedule directly within the event platform, once live early July 2021. All times are listed in Pacific Time.

Machine Learning Summit: Deep Reinforcement Learning For Navigation

Maxim Peter  (R&D Programmer, Ubisoft)

Joshua Romoff  (R&D Scientist, Ubisoft)

Date: Tuesday, July 20

Time: 1:20pm - 1:50pm

Pass Type: All Access Pass, Summits Pass

Topic: Programming

Format: Session

Vault Recording: Video

Audience Level: N/A: I prefer to present live on-site but I am open to delivering a virtual presentation

Navigation is at the heart of every AI behaviors in games. However, existing navigation solutions such as Navigation Meshes do not scale well when characters can use complex (yet common in games) navigation abilities, such as being able to jump several times, using a jetpack, teleporting, wall jumping, using jump pads ... In this talk, Maxim Peter and Joshua Romoff will dig into the details of a novel alternative explored at Ubisoft La Forge, proposing to use Deep Reinforcement Learning (Deep RL) to teach bots interacting directly with a game pad to navigate on complex maps using navigation abilities. They will explain how this solution works, discuss the advantages and limits of their current solution and provide examples of successfull integrations in Ubisoft games.


Attendees will leave with a better understanding of how Deep Reinforcement Learning can be used for navigation in games. They will hear about concrete examples and experiences, as well as our vision for its future development.

Intended Audience

AI and gameplay programmers, as well as Machine Learning and Reinforcement Learning enthusiasts should be interested in this presentation. While no prerequisites are required to follow the talk, an understanding of graph-based navigation techniques and Machine Learning would be a plus.