Updated
Updated · HPCwire · Sep 3
JPMorganChase, Argonne Unveil QAOA Simulation Method for 1 Large-System Limit
Updated
Updated · HPCwire · Sep 3

JPMorganChase, Argonne Unveil QAOA Simulation Method for 1 Large-System Limit

3 articles · Updated · HPCwire · Sep 3

Summary

  • Researchers from JPMorganChase and Argonne developed a technique that estimates QAOA solution quality on large optimization problems without running the full quantum algorithm end to end.
  • The method maps the high-depth, infinite-size Sherrington-Kirkpatrick model to a single quantum spin coupled to bosonic modes, replacing costly calculations with simpler matrix product state simulations.
  • Argonne's Jeffrey Larson also built an optimization approach to find strong QAOA parameters, with simulations run on ALCF and NERSC supercomputers through a DOE INCITE grant.
  • Published in Physical Review Letters, the work aims to clarify how far quantum optimization algorithms can scale as hardware matures and whether claimed quantum advantage is achievable on larger systems.

Insights

Why are supercomputers still essential to test QAOA, even when the goal is to prove quantum optimization can beat classical methods?
If QAOA for a hard spin-glass can be classically estimated through a spin-boson shortcut, where does true quantum advantage still survive?
Can a high-depth mapping from QAOA to a simple spin-boson model predict real hardware performance, or only ideal infinite-size behavior?