Minecraft and Reinforcement Learning
2 likes350 views
The document discusses the integration of Minecraft with reinforcement learning through Project Malmo, an open-source platform developed by Microsoft Research for AI experimentation. It explains key concepts of reinforcement learning, including trial-and-error learning, and highlights technical parameters such as Q-learning and deep reinforcement learning. Additionally, it provides resource links and mentions the proprietary nature of the content and associated software.
Minecraft and Reinforcement Learning
- 1. Lars Gregori, SAP Hybris Labs April 13, 2018 Minecraft and Reinforcement Learning AI Hacking
- 2. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ
- 3. Minecraft
- 4. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ #AIandMinecraft @choas Minecraft Markus "Notch" Persson Mojang AB Best-selling PC game of all time Exploration Resource gathering Crafting Combat Sandbox construction game Creative + building aspects Three-dimensional environment
- 6. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ #AIandMinecraft @choas Project Malmo Open Source (Github) Microsoft Research Lab Based on Minecraft / Minecraft Forge Agents written in Python, Lua, C++, C#, Java, Torch, ALE Mission XML WorldState Send Command
- 8. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ #AIandMinecraft @choas Reinforcement Learning Supervised Learning Unsupervised Learning Reinforcement Learning
- 9. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ #AIandMinecraft @choas Reinforcement Learning Observation Reward Action Environment Agent
- 10. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ #AIandMinecraft @choas “Reinforcement learning is like trial-and-error learning.”David Silver Reinforcement Learning
- 11. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ Public "11 Reinforcement Learning: An Introduction Richard S. Sutton and Andrew G. Barto (1998) Reinforcement Learning
- 12. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ Reinforcement Learning Example
- 13. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ Public "13 Q-Learning
- 14. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ Public "14 Q-Learning 100 -1 99.0 α - step-size parameter (1.0) γ - discount-rate parameter (0.8)
- 15. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ Public "15 Q-Learning 100 -1 99.0 78.2 -101.0 α - step-size parameter (1.0) γ - discount-rate parameter (0.8) -100 -1
- 16. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ Public "16 Q-Learning 99.0000000 78.2000000 61.5600000 48.2480000 37.5984000 29.0787200 22.2629760 16.8103808 12.4483046 8.9586437 99 78 61 48 37 29 2216 128
- 18. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ #AIandMinecraft @choas Deep Reinforcement Learning Supervised Learning Unsupervised Learning Reinforcement Learning
- 19. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ #AIandMinecraft @choas Deep Reinforcement Learning https://youtu.be/TmPfTpjtdgg
- 20. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ
- 21. Thank you. Contact information: Lars Gregori @choas
- 22. PUBLIC© 2017 SAP SE or an SAP affiliate company. All rights reserved. ǀ #AIandMinecraft @choas The Malmo Platform for Artificial Intelligence Experimentation. Proc. 25th International Joint Conference on Artificial Intelligence http://www.ijcai.org/Proceedings/2016 Project Malmo https://www.microsoft.com/en-us/research/project/project-malmo/ Project Malmo (Github) https://github.com/Microsoft/malmo Reinforcement Learning: An Introduction - ISBN-13: 978-0262193986 2nd Version online YouTube RL Course by David Silver Links
- 23. No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP SE or an SAP affiliate company. The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its distributors contain proprietary software components of other software vendors. National product specifications may vary. These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials. The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional warranty. In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or any related presentation, or to develop or release any functionality mentioned therein. This document, or any related presentation, and SAP SE’s or its affiliated companies’ strategy and possible future developments, products, and/ or platform directions and functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time for any reason without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or functionality. All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ materially from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, and they should not be relied upon in making purchasing decisions. SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names mentioned are the trademarks of their respective companies. See http://global.sap.com/corporate-en/legal/copyright/index.epx for additional trademark information and notices. © 2018 SAP SE or an SAP affiliate company. All rights reserved.