Sample Efficient Reinforcement Learning through Learning from Demonstrations in Minecraft

2020/03/12 by Scheller, Christian, Schraner, Yanick, Vogel, Manfred
#FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.2003.06066

Abstract

Sample inefficiency of deep reinforcement learning methods is a major obstacle for their use in real-world applications. In this work, we show how human demonstrations can improve final performance of agents on the Minecraft minigame ObtainDiamond with only 8M frames of environment interaction. We propose a training procedure where policy networks are first trained on human data and later fine-tuned by reinforcement learning. Using a policy exploitation mechanism, experience replay and an additional loss against catastrophic forgetting, our best agent was able to achieve a mean score of 48. Our proposed solution placed 3rd in the NeurIPS MineRL Competition for Sample-Efficient Reinforcement Learning.

Related

• The MineRL 2019 Competition on Sample Efficient Reinforcement Learning using Human Priors
• JueWu-MC: Playing Minecraft with Sample-efficient Hierarchical Reinforcement Learning
• Forgetful Experience Replay in Hierarchical Reinforcement Learning from Demonstrations
• The MineRL 2020 Competition on Sample Efficient Reinforcement Learning using Human Priors
• MineRL: A Large-Scale Dataset of Minecraft Demonstrations