Quantum Circuit Architecture Search on a Superconducting Processor
Kehuan Linghu,
Yang Qian,
Ruixia Wang,
Meng-Jun Hu,
Zhiyuan Li,
Xuegang Li,
Huikai Xu,
Jingning Zhang,
Teng Ma,
Peng Zhao,
Dong E. Liu,
Min-Hsiu Hsieh,
Xingyao Wu,
Yuxuan Du,
Dacheng Tao,
Yirong Jin,
Haifeng Yu
Affiliations
Kehuan Linghu
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Yang Qian
School of Computer Science, Faculty of Engineering, University of Sydney, Camperdown, NSW 2006, Australia
Ruixia Wang
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Meng-Jun Hu
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Zhiyuan Li
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Xuegang Li
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Huikai Xu
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Jingning Zhang
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Teng Ma
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Peng Zhao
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Dong E. Liu
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Min-Hsiu Hsieh
Centre for Quantum Software and Information, Faculty of Engineering and Information Technology, University of Technology Sydney, Ultimo, NSW 2007, Australia
Xingyao Wu
JD Explore Academy, Beijing 102628, China
Yuxuan Du
JD Explore Academy, Beijing 102628, China
Dacheng Tao
JD Explore Academy, Beijing 102628, China
Yirong Jin
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Haifeng Yu
Beijing Academy of Quantum Information Sciences, Beijing 100193, China
Variational quantum algorithms (VQAs) have shown strong evidence to gain provable computational advantages in diverse fields such as finance, machine learning, and chemistry. However, the heuristic ansatz exploited in modern VQAs is incapable of balancing the trade-off between expressivity and trainability, which may lead to degraded performance when executed on noisy intermediate-scale quantum (NISQ) machines. To address this issue, here, we demonstrate the first proof-of-principle experiment of applying an efficient automatic ansatz design technique, i.e., quantum architecture search (QAS), to enhance VQAs on an 8-qubit superconducting quantum processor. In particular, we apply QAS to tailor the hardware-efficient ansatz toward classification tasks. Compared with heuristic ansätze, the ansatz designed by QAS improves the test accuracy from 31% to 98%. We further explain this superior performance by visualizing the loss landscape and analyzing effective parameters of all ansätze. Our work provides concrete guidance for developing variable ansätze to tackle various large-scale quantum learning problems with advantages.
