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.
Linxia Gong (PI of Matchmaking R&D at Fuxi AI Lab, Netease)
Date: Thursday, July 22
Time: 4:20pm - 5:20pm
Pass Type: All Access Pass, Core Pass
Topic: Design, Programming
Vault Recording: Video
Audience Level: N/A: I prefer to present live on-site but I am open to delivering a virtual presentation
Matchmaking is crucial for the player satisfaction in the online player-versus-player games. Many factors should be considered in a matchmaking system design, such as waiting time in the matchmaking queue, network latency difference, skill gaps, defeat streaks, etc. We present the findings and algorithms of our matchmaking practice in Netease Games, about how to model the effects of those factors and optimize the gross players' utility in the assignment process:
1. win patterns: the factors that influence the victory in a match, and the algorithms that incorporate those factors during the application of machine learning
2. churn patterns: the factors that influence the players' churn after this matchmaking proposal, and the tricks to optimize the matchmaking process
3. the matchmaking planning algorithms, which is about solving a combinatorial optimization problem
4. the matchmaking Saas practice in Netease Games.
The attendees will learn the win patterns and churn patterns which should be considered in matchmaking process, how to apply AI methods to better depict heroes, players and game modes, and how to optimize the matchmaking process by solving the combinatorial optimization problem. And the attendees will learn how to design and apply a matchmaking Saas.
Programmers and designers of all experience levels will be interested in applying AI to understand their game environments and players, and will find operationally-efficient ways to optimize the group competition matchmaking process.
Existing AI practitioners will be inspired with the new applications of well understood techniques.