Guided sampling ans\"atzes for variational quantum computing
AI Breakdown
Get a structured breakdown of this paper — what it's about, the core idea, and key takeaways for the field.
Abstract
Quantum computing is a promising technology because of the ability of quantum computers to process vector spaces with dimensions that increase exponentially with the simulated system size. Extracting the solution, however, is challenging as the number of quantum gate operations and quantum circuit executions must still scale at most polynomially. Consequently, choosing a good ansatz--a polynomial subset of the exponentially many possible solutions--will be critical to maintain accuracy for larger systems. To address this challenge, we introduce a class of guided sampling ans\"atzes (GSAs) that depend on the system interactions and measured state samples as well as a parameter space. We demonstrate a minimal ansatz for the hydronium cation H$_3$O$^+$ and found that with only 200 circuit executions per structure on the IonQ Aria quantum computer, our calculations produced total energies around the relaxed structure with errors well below $1.59\times10^{-3}$ Ha, thus exceeding chemical accuracy.