Illuminating Generalization in Deep Reinforcement Learning through Procedural Level Generation
This addresses the generalization issue in deep RL for video games, which is incremental as it builds on existing methods with procedural generation.
The paper tackles the problem of deep reinforcement learning agents overfitting to fixed environments by using procedural level generation during training, showing that it enables generalization to new levels within the same distribution and can achieve better performance with less data through difficulty manipulation.
Deep reinforcement learning (RL) has shown impressive results in a variety of domains, learning directly from high-dimensional sensory streams. However, when neural networks are trained in a fixed environment, such as a single level in a video game, they will usually overfit and fail to generalize to new levels. When RL models overfit, even slight modifications to the environment can result in poor agent performance. This paper explores how procedurally generated levels during training can increase generality. We show that for some games procedural level generation enables generalization to new levels within the same distribution. Additionally, it is possible to achieve better performance with less data by manipulating the difficulty of the levels in response to the performance of the agent. The generality of the learned behaviors is also evaluated on a set of human-designed levels. The results suggest that the ability to generalize to human-designed levels highly depends on the design of the level generators. We apply dimensionality reduction and clustering techniques to visualize the generators' distributions of levels and analyze to what degree they can produce levels similar to those designed by a human.