Playing Minecraft with Behavioural Cloning

Abstract

MineRL 2019 competition challenged participants to train sample-efficient agents to play Minecraft, by using a dataset of human gameplay and a limit number of steps the environment. We approached this task with behavioural cloning by predicting what actions human players would take, and reached fifth place in the final ranking. Despite being a simple algorithm, we observed the performance of such an approach can vary significantly, based on when the training is stopped. In this paper, we detail our submission to the competition, run further experiments to study how performance varied over training and study how different engineering decisions affected these results.

arXiv commentTo appear in Post Proceedings of the Competitions & Demonstrations Track @ NeurIPS2019. Source code available at https://github.com/Miffyli/minecraft-bc

References in corpus (9)

• SMOTE: Synthetic Minority Over-sampling Technique
• End to End Learning for Self-Driving Cars
• IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
• Deep Reinforcement Learning that Matters
• Deep Q-learning from Demonstrations
• Benchmarking Batch Deep Reinforcement Learning Algorithms
• Learning to Act by Predicting the Future
• The Multi-Agent Reinforcement Learning in MalmÖ (MARLÖ) Competition
• The MineRL 2019 Competition on Sample Efficient Reinforcement Learning using Human Priors

Cited by in corpus (1)

• Retrospective Analysis of the 2019 MineRL Competition on Sample Efficient Reinforcement Learning