【Date and Time】26-June-2026 2:00pm (Beijing time)

【Venue】Room 320

【Host】Mengjun Hu (BAQIS)

【Language】English

【Title】Demonstration of Efficient Predictive Surrogates for Large-scale Quantum Processors

【Speaker】 

Xinbiao Wang is a postdoctoral research fellow at Nanyang Technological University, working with Prof. Dacheng Tao and Prof. Yuxuan Du. He obtained his Ph.D. from Wuhan University in 2024. His research interests include quantum machine learning, quantum learning theory, and AI for quantum science. He has published his work in leading journals and conferences in physics and AI, including Nature Communications, Quantum, TNNLS, NeurIPS, ICLR, and AAAI.

【Abstract】

The ongoing development of quantum processors is driving breakthroughs in scientific discovery. Despite this progress, the formidable cost of fabricating large-scale quantum processors means they will remain rare for the foreseeable future, limiting their widespread application. In this talk, I will show how classical learning models can address this bottleneck. These models, which we call predictive surrogates, are designed to emulate the mean-value behavior of a given quantum processor with provable computational efficiency. I will present two such surrogates and show how they substantially reduce the need for quantum processor access across a range of applications, including ground state energy estimation via variational quantum eigensolvers (VQE) and quantum phase identification. To illustrate their potential, I will walk through experiments on up to 20 superconducting qubits, where we use these surrogates to pre-train VQEs for transverse-field Ising models and to identify Floquet symmetry-protected topological phases. I will then discuss our key finding: the predictive surrogates reduce measurement overhead by orders of magnitude and, in some cases, surpass conventional quantum-resource-intensive approaches. Collectively, these findings establish predictive surrogates as a practical pathway to broadening the impact of advanced quantum processors.