The Atari Grand Challenge Dataset

Abstract

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.

References in corpus (6)

• Adam: A Method for Stochastic Optimization
• Generative Adversarial Imitation Learning
• Playing Atari Games with Deep Reinforcement Learning and Human Checkpoint Replay
• A Deep Learning Approach for Joint Video Frame and Reward Prediction in Atari Games
• Learning to Repeat: Fine Grained Action Repetition for Deep Reinforcement Learning
• Model-based Adversarial Imitation Learning

Cited by in corpus (12)

• Inverse reinforcement learning for video games
• Atari-HEAD: Atari Human Eye-Tracking and Demonstration Dataset
• A novel policy for pre-trained Deep Reinforcement Learning for Speech Emotion Recognition
• Deep Reinforcement Learning with Pre-training for Time-efficient Training of Automatic Speech Recognition
• AI Enabling Technologies: A Survey
• MineRL: A Large-Scale Dataset of Minecraft Demonstrations
• Pre-training in Deep Reinforcement Learning for Automatic Speech Recognition
• Learning Calibratable Policies using Programmatic Style-Consistency
• Reinforcement Learning Under Algorithmic Triage
• Constrained Policy Improvement for Safe and Efficient Reinforcement Learning
• Parallelized Interactive Machine Learning on Autonomous Vehicles
• Adaptive Agent Architecture for Real-time Human-Agent Teaming