Contrastive Learning of Generalized Game Representations

18 Jun 2021  Â·  Chintan Trivedi, Antonios Liapis, Georgios N. Yannakakis ·

Representing games through their pixels offers a promising approach for building general-purpose and versatile game models. While games are not merely images, neural network models trained on game pixels often capture differences of the visual style of the image rather than the content of the game. As a result, such models cannot generalize well even within similar games of the same genre. In this paper we build on recent advances in contrastive learning and showcase its benefits for representation learning in games. Learning to contrast images of games not only classifies games in a more efficient manner; it also yields models that separate games in a more meaningful fashion by ignoring the visual style and focusing, instead, on their content. Our results in a large dataset of sports video games containing 100k images across 175 games and 10 game genres suggest that contrastive learning is better suited for learning generalized game representations compared to conventional supervised learning. The findings of this study bring us closer to universal visual encoders for games that can be reused across previously unseen games without requiring retraining or fine-tuning.

PDF Abstract

Datasets


Introduced in the Paper:

Sports10

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Classification Sports10 Max Margin Contrastive Validation Accuracy 93.42 # 1
Representation Learning Sports10 Max Margin Contrastive Silhouette Score 0.56 # 1

Methods