Speaker: Bijita Sarma (OIST, Okinawa)

Title: Deep reinforcement learning for measurement-based feedback quantum control and machine-optimized pulsed cooling

Abstract: In this talk, I will use two of our recent works [1, 2] to show how Deep Reinforcement Learning (DRL) can be effectively used to develop non-trivial controls for quantum systems. In the first part of the talk, I will show feedback control of a nonlinear system with DRL as the controller, trained on noisy measurement data obtained from weak continuous measurements [1]. In the second part, I will show how DRL can find non-intuitive pulse sequences suitable for open-loop control to effectively cool a `hot’ mode through bi/tri-partite interactions within and beyond regimes of rotating wave approximation [2].

References:

[1] Sangkha Borah, Bijita Sarma, Michael Kewming, Gerard J. Milburn, and Jason Twamley, “Measurement-Based Feedback Quantum Control with Deep Reinforcement Learning for a Double-Well Nonlinear Potential,” Physical Review Letters 127, 19 (2021): 190403. https://doi.org/10.1103/PhysRevLett.127.190403

[2] Bijita Sarma, Sangkha Borah, A. Kani, and Jason Twamley, “Accelerated Magnonic Motional Cooling with Deep Reinforcement Learning,” arXiv:2204.07710 [Quant-Ph], April 15, 2022. http://arxiv.org/abs/2204.07710