Machine-learned trust scoring for player matchmaking

摘要

The trained machine learning model(s) is used to determine a score (e.g., a trust score) for a user account registered with the video game service, and the score is used to match players together in a multiplayer video game setup. do. In an exemplary process, the computing system can access data associated with a registered user account and provide data as input to the trained machine learning model(s), where the trained machine learning model(s) is the player can multiplayer. While playing a video game in a mod, a score related to the probability of acting or not acting according to a specific action is generated as an output. Thereafter, a subset of the logged-in user accounts that play the video game may be assigned to different matches based at least in part on the score determined for the logged-in user account, and the video game is assigned for each logged-in user account. It is executed in the game.