Experimentally realizing efficient quantum control with reinforcement learning

• Published in: Science China. Physics, mechanics & astronomy Vol. 65; no. 5; p. 250312
• Main Authors: Ai, Ming-Zhong, Ding, Yongcheng, Ban, Yue, Martín-Guerrero, José D., Casanova, Jorge, Cui, Jin-Ming, Huang, Yun-Feng, Chen, Xi, Li, Chuan-Feng, Guo, Guang-Can
• Format: Journal Article
• Language: English
• Published: Beijing Science China Press 01.05.2022; Springer; Springer Nature B.V
• Subjects: Analysis; Artificial intelligence; Astronomy; Classical and Continuum Physics; Cost analysis; Deep learning; Energy costs; Errors; Mathematical analysis; Observations and Techniques; Physics; Physics and Astronomy; Quantum theory; Qubits (quantum computing); Robust control; trapped ion; quantum control; quantum computing; noise robustness; reinforcement learning
• ISSN: 1674-7348; 1869-1927
• DOI: 10.1007/s11433-021-1841-2

We experimentally investigate deep reinforcement learning (DRL) as an artificial intelligence approach to control a quantum system. We verify that DRL explores fast and robust digital quantum controls with operation time analytically hinted by shortcuts to adiabaticity. In particular, the protocol’s robustness against both over-rotations and off-resonance errors can still be achieved simultaneously without any priori input. For the thorough comparison, we choose the task as single-qubit flipping, in which various analytical methods are well-developed as the benchmark, ensuring their feasibility in the quantum system as well. Consequently, a gate operation is demonstrated on a trapped 171 Yb + ion, significantly outperforming analytical pulses in the gate time and energy cost with hybrid robustness, as well as the fidelity after repetitive operations under time-varying stochastic errors. Our experiments reveal a framework of computer-inspired quantum control, which can be extended to other complicated tasks without loss of generality.