Deep learning for data-driven skill estimation

1. A computer-implemented method for determining a player's skill for video games, comprising:

aggregating a plurality of player statistics for match outcomes from a plurality of different video games;

calculating, for each player in a pool of players, a matchmaking rating for each player based on the plurality of player statistics, the matchmaking rating for each player comprising a predicted number of points each player will contribute to a team point total of a match;

selecting, based on the matchmaking rating for each player, players from the pool of players, wherein the pool of players are requesting to participate in the match;

matching the players in the pool of players that requested to join the match based on the matchmaking rating for each player, a sum of the matchmaking ratings of the matched players comprising a total predicted team score for the match; and

training a machine learning algorithm based on the plurality of player statistics to determine a correlation between the plurality of player statistics and the match outcomes for at least one video game of the plurality of different video games, the machine learning algorithm comprising a Siamese neural network.

2. The computer-implemented method of claim 1 , further comprising:

receiving requests for matchmaking from a plurality of players.

3. The computer-implemented method of claim 1 , wherein the Siamese neural network comprises a branch for each player.

4. The computer-implemented method of claim 1 , wherein the training is based at least in part on contextual data comprising at least one of in-game maps, game modes, and/or player roles.

5. The computer-implemented method of claim 4 , wherein the matching is based at least in part on the contextual data.

6. The computer-implemented method of claim 1 , wherein the matching is dynamic.

7. The computer-implemented method of claim 1 , wherein the match comprises at least a first team playing against a second team, and a predicted score for each team comprises a sum of the matchmaking ratings of all the players for that team.

8. The computer-implemented method of claim 1 , wherein the matching is based at least in part on player preferences.

9. A system for determining a player's skill for video games, comprising:

a processor; and

a memory comprising instructions stored thereon, which when executed by the processor, causes the processor to perform:

aggregating a plurality of player statistics for match outcomes from a plurality of different video games;

calculating, for each player in a pool of players, a matchmaking rating for each player based on the plurality of player statistics, the matchmaking rating for each player comprising a predicted number of points each player will contribute to a team point total of a match;

selecting, based on the matchmaking rating for each player, players from the pool of players, wherein the pool of players are requesting to participate in the match;

matching the players in the pool of players that requested to join the match based on the matchmaking rating for each player, a sum of the matchmaking ratings of the matched players comprising a total predicted team score for the match; and

training a machine learning algorithm based on the plurality of player statistics to determine a correlation between the plurality of player statistics and the match outcomes for at least one video game of the plurality of different video games, the machine learning algorithm comprising a Siamese neural network.

10. The system of claim 9 , further comprising stored sequences of instructions, which when executed by the processor, cause the processor to perform:

receiving requests for matchmaking from a plurality of players.

11. The system of claim 9 , wherein the Siamese neural network comprises a branch for each player.

12. The system of claim 9 , wherein the training is based at least in part on contextual data comprising at least one of in-game maps, game modes, and/or player roles.

13. The system of claim 12 , wherein the matching is based at least in part on the contextual data.

14. The system of claim 9 , wherein the match comprises at least a first team playing against a second team, and a predicted score for each team comprises a sum of the matchmaking ratings of all the players for that team.

15. The system of claim 9 , wherein the matching is based at least in part on player preferences.

16. A non-transitory computer-readable storage medium comprising instructions stored thereon, which when executed by one or more processors, cause the one or more processors to perform operations for determining a player's skill for video games, the operations comprising:

aggregating a plurality of player statistics for match outcomes from a plurality of different video games;

calculating, for each player in a pool of players, a matchmaking rating for each player based on the plurality of player statistics, the matchmaking rating for each player comprising a predicted number of points each player will contribute to a team point total of a match;

selecting, based on the matchmaking rating for each player, players from the pool of players, wherein the pool of players are requesting to participate in the match;

matching the players in the pool of players that requested to join the match based on the matchmaking rating for each player, a sum of the matchmaking ratings of the matched players comprising a total predicted team score for the match; and

training a machine learning algorithm based on the plurality of player statistics to determine a correlation between the plurality of player statistics and the match outcomes for at least one video game of the plurality of different video games, the machine learning algorithm comprising a Siamese neural network.