The Atari Grand Challenge Dataset
This dataset addresses a key bottleneck for researchers in reinforcement learning by providing accessible human demonstration data, though it is incremental as it builds on existing imitation learning approaches.
The authors tackled the scarcity of human demonstration data for reinforcement learning by collecting the largest and diverse dataset of human Atari 2600 replays, and they demonstrated its use by analyzing how demonstration quality affects imitation learning performance.
Recent progress in Reinforcement Learning (RL), fueled by its combination, with Deep Learning has enabled impressive results in learning to interact with complex virtual environments, yet real-world applications of RL are still scarce. A key limitation is data efficiency, with current state-of-the-art approaches requiring millions of training samples. A promising way to tackle this problem is to augment RL with learning from human demonstrations. However, human demonstration data is not yet readily available. This hinders progress in this direction. The present work addresses this problem as follows. We (i) collect and describe a large dataset of human Atari 2600 replays -- the largest and most diverse such data set publicly released to date, (ii) illustrate an example use of this dataset by analyzing the relation between demonstration quality and imitation learning performance, and (iii) outline possible research directions that are opened up by our work.