Probability distribution analysis of the cascaded variational quantum eigensolver

Abstract:The cascaded variational quantum eigensolver (CVQE) circumvents the need for iterative communication between the quantum and classical processing units that is necessary in the conventional VQE algorithm. While CVQE offers complete freedom to choose the guiding state as input, not all guiding states provide both accurate solutions and efficient resource usage. Our work presents a process based on trapezoidal-state preparation for selecting guiding states that yield accurate many-electron ground-state solutions with minimal resource consumption. By analyzing state probability distributions at different stages of CVQE calculations for the collision complex mediating the $\mathrm{H}_2 + \mathrm{H}_2^+ \rightarrow \mathrm{H}_3^+ + \mathrm{H}$ bimolecular reaction, we identify stage-specific errors, trace their impact on final classical optimization, and determine the optimal guiding-state parameters for given resource constraints. Our findings uncover a critical trade-off on noisy intermediate-scale quantum (NISQ) devices, demonstrating that shallower diabatic circuits provide the best compromise among optimization efficiency, state quality, sampling cost, and signal-to-noise ratio in our system.

arXiv:2605.00807 [quant-ph]

(or arXiv:2605.00807v2 [quant-ph] for this version)

https://doi.org/10.48550/arXiv.2605.00807 Focus to learn more arXiv-issued DOI via DataCite

Phys. Rev. A 114, 042436 (2026)

https://doi.org/10.1103/h17r-4ldq Focus to learn more DOI(s) linking to related resources

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.