Griddly: A platform for AI research in games
This provides a tool for AI researchers to test reinforcement learning agents across varied environments, addressing overfitting and generalization issues, though it is incremental as it builds on existing platform concepts.
The paper tackles the difficulty of prototyping diverse game environments for AI research by introducing Griddly, a platform that offers configurable games and efficient performance, with baseline experiments showing effects on RL agent generalization.
In recent years, there have been immense breakthroughs in Game AI research, particularly with Reinforcement Learning (RL). Despite their success, the underlying games are usually implemented with their own preset environments and game mechanics, thus making it difficult for researchers to prototype different game environments. However, testing the RL agents against a variety of game environments is critical for recent effort to study generalization in RL and avoid the problem of overfitting that may otherwise occur. In this paper, we present Griddly as a new platform for Game AI research that provides a unique combination of highly configurable games, different observer types and an efficient C++ core engine. Additionally, we present a series of baseline experiments to study the effect of different observation configurations and generalization ability of RL agents.