Location: Zoom https://harvard.zoom.us/j/779283357?pwd=MitXVm1pYUlJVzZqT3lwV2pCT1ZUQT09
Time: Tuesday, May 11, 2021, 10:00 AM (Eastern US), 15:00 (UK/Eire), 16:00 (Central Europe), 22:00 (China)
Title: Understanding the Linear Cross Entropy Benchmark through a Statistical Physics Model
Abstract: In a recent experiment, Google claimed to achieve quantum advantage based on sampling from random quantum circuits. To certify the correctness of the sampling task, they developed a linear cross- entropy benchmark (XEB) and made their claim based on achieving a high value of XEB. The implicit assumption made in Google’s claim is that XEB approximates the fidelity well and hence could serve as a proxy in practice. However, it turns out that XEB can often deviate from fidelity, so it is crucial to understand how XEB and fidelity are related. I introduce a novel approach to analyze XEB through a mapping of random quantum circuits to a discrete classical statistical-physics diffusion-reaction model. By analyzing this model both qualitatively and numerically, we show that XEB usually overestimates fidelity These results indicate that one needs a new benchmark to certify quantum advantage.
Additional Ways to Join
Join by telephone (use any number to dial in)
+1 929 436 2866
+1 312 626 6799
+1 669 900 6833
400 669 9381 China Toll-free
International numbers available: https://harvard.zoom.us/u/aclg6kOggb
One tap mobile: +19294362866,,779283357# US (New York)
Join by SIP conference room system
Meeting ID: 779 283 357
779283357@zoomcrc.com