An adaptive deep reinforcement learning framework enables curling robots with human-like performance in real-world conditions-Reference-Cited by-同舟云学术

An adaptive deep reinforcement learning framework enables curling robots with human-like performance in real-world conditions

Published:2020-09-30 Issue:46 Volume:5 Page:
ISSN:2470-9476
Container-title:Science Robotics
language:en
Short-container-title:Sci. Robot.

Author:

Won Dong-Ok¹^ORCID,Müller Klaus-Robert¹²³^ORCID,Lee Seong-Whan¹⁴^ORCID

Affiliation:

1. Department of Brain and Cognitive Engineering, Korea University, Seoul, Republic of Korea.

2. Machine Learning Group, Department of Computer Science, Berlin Institute of Technology, Berlin, Germany.

3. Max Planck Institute for Informatics, Saarbrücken, Germany.

4. Department of Artificial Intelligence, Korea University, Seoul, Republic of Korea.

Abstract

An AI curling robot can compete and win against top-ranked human teams in the game of curling.

Funder

German ministry for education and research

Institute for Information and Communications Technology Planning and Evaluation (IITP) grant funded by the Korea government

Microsoft Research Asia

Publisher

American Association for the Advancement of Science (AAAS)

Subject

Artificial Intelligence,Control and Optimization,Computer Science Applications,Mechanical Engineering

Reference79 articles.

1. Learning, planning, and control for quadruped locomotion over challenging terrain

2. T. Yee V. Lisý M. H. Bowling Monte Carlo tree search in continuous action spaces with execution uncertainty in Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI) (AAAI Press 2016) pp. 690–697.

3. Z. F. Ahmad R. C. Holte M. Bowling Action selection for hammer shots in curling in Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI) (AAAI Press 2016) pp. 561–567.

4. J. Tobin R. Fong A. Ray J. Schneider W. Zaremba P. Abbeel Domain randomization for transferring deep neural networks from simulation to the real world in Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (24 to 28 September 2017) pp. 23–30.

5. J. Tan T. Zhang E. Coumans A. Iscen Y. Bai D. Hafner S. Bohez V. Vanhoucke Sim-to-real: Learning agile locomotion for quadruped robots. arXiv:1804.10332 [cs.RO] (27 April 2018).

Cited by 38 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. An intelligent dual-sensing e-skin system for pressure and temperature detection using laser-induced graphene and polydimethylsiloxane;Materials & Design;2024-02

2. Personalized robotic control via constrained multi-objective reinforcement learning;Neurocomputing;2024-01

3. A mediation analysis of the motivation for learning in the process of knowledge acquisition and transfer;The Learning Organization;2023-12-01

4. High-dimensional reinforcement learning for optimization and control of ultracold quantum gases;Machine Learning: Science and Technology;2023-12-01

5. QvQ-IL: quantity versus quality in incremental learning;Neural Computing and Applications;2023-11-24