aiqbench analysis
This study demonstrates that classical tensor network simulations can efficiently reproduce results from quantum hardware experiments on the Fermi-Hubbard model, challenging the immediate necessity of quantum processors for specific near-term benchmarks. By achieving higher accuracy and longer time-scale predictions with modest computational resources, the research highlights a critical boundary where classical methods remain superior or competitive against emerging digital quantum advantages.
Fast classical simulation of `Fast, accurate, high-resolution simulation of large-scale Fermi-Hubbard models on a digital quantum processor'
arXiv:2608.13805v1 Announce Type: new Abstract: We study the N\'{e}el quench dynamics of a 1D Fermi-Hubbard model which has recently been simulated on quantum hardware. We demonstrate that the set of 7260 observable trajectories measured in the quantum experiment can be obtained more quickly and accurately through classical tensor network simulation using modest computation. Our result relies on transverse tensor network contraction, where a bond dimension of 32 is already sufficient to reproduce the quantum experiment. We further extend the converged observable trajectories to longer times than in the hardware simulation and in other recent classical simulations.
Abstract:We study the Néel quench dynamics of a 1D Fermi-Hubbard model which has recently been simulated on quantum hardware. We demonstrate that the set of 7260 observable trajectories measured in the quantum experiment can be obtained more quickly and accurately through classical tensor network simulation using modest computation. Our result relies on transverse tensor network contraction, where a bond dimension of 32 is already sufficient to reproduce the quantum experiment. We further extend the converged observable trajectories to longer times than in the hardware simulation and in other recent classical simulations.
arXiv:2608.13805 [quant-ph]
(or arXiv:2608.13805v1 [quant-ph] for this version)
https://doi.org/10.48550/arXiv.2608.13805 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 182 words
