Tianyi Hao

Title: Enabling High Performance Debugging for Variational Quantum Algorithms using Compressed Sensing

Abstract: Variational quantum algorithms (VQAs) are promising for solving practical problems on Noisy Intermediate Scale Quantum (NISQ) devices. However, developing VQAs is challenging due to the limited availability of quantum hardware, their high error rates, and the significant overhead of classical simulations. In addition, for a VQA to work, researchers must choose proper configurations for its various components in an empirical manner, as there are few techniques or software tools to configure and tune the VQA hyperparameters.In this talk, Tianyi will present OSCAR, a tool to help VQA researchers pick the right initial points, choose suitable optimizer configurations, and deploy appropriate error mitigation methods. OSCAR enables efficient debugging and performance tuning by providing users with the loss function landscape without running thousands of quantum circuits as required by the grid search. Furthermore, OSCAR can compute an optimizer function query in an instant by interpolating a computed landscape, thus significantly reducing the overhead of optimizer hyperparameter tuning.

Bio: Tianyi Hao is a CS Ph.D. student at the University of Wisconsin-Madison in Prof. Swamit Tannu’s group. Prior to joining UW-Madison, he received his Bachelor’s degrees in CS and Physics at the University of Illinois at Urbana-Champaign and Master’s degree in CS at Stanford. His research interests include quantum algorithms, quantum optimization, and quantum circuit simulation.

Contact the speaker: