Quantum reinforcement learning

Shao hua Wang, Hai Lin

Keywords: quantum reinforcement learning, multi-agent, quantum technology, control optimization, quantum circuit

TL;DR: A review and implementation of quantum reinforcement learning. We used QRL to train several game agents, and finally predicted and looked forward to future applications and trends.

Abstract: With the rapid development of quantum technology, it has been confirmed that it can surpass the speed of traditional computing in some fields. Quantum advantage can also be manifested in the field of machine learning. We reviewed many current papers related to quantum reinforcement learning. We discuss in depth how quantum reinforcement learning is implemented and core techniques. quantum reinforcement learning (QRL) method is proposed by combining quantum theory and reinforcement learning (RL).The field of quantum reinforcement learning actually includes two aspects: One is use quantum properties to help reinforcement learning, the other is using reinforcement learning to help quantum circuit design. We have completed agent training for several classic games using quantum reinforcement learning methods, and the superiority and feasibility of the simulation experiments were evaluated. The QRL algorithm can be used in many aspects such as finance, industrial simulation, mechanical control, quantum communication, and quantum circuit optimization. We take a look at the field of quantum reinforcement learning and make bold predictions that many applications in the future will benefit from the development of this technology.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)