Applied Physics Seminar

In person and via webinar!

NOTE: At this time, in-person APhMS seminars are open to all Caltech students/staff/faculty/visitors with a valid Caltech ID. Outside community members are welcome to join our online event.

Webinar Link: https://caltech.zoom.us/j/84603560086

Webinar ID: 846 0356 0086

Abstract:

The electronic structure (ES) problem is an important application for a quantum computer, and recent years have witnessed the emergence of the first quantum algorithms for ES simulations [1].

Notwithstanding this progress, due to the immaturity of contemporary quantum computation platforms, so far quantum ES simulations have only provided conceptually interesting results. Progress in increasing the relevancy of quantum ES simulations requires the concerted development of algorithms tailored for contemporary quantum devices, and error mitigation techniques. This contribution will describe examples of both methodologies.

A challenge in quantum ES simulations is the high cost of performing Trotter steps of time evolution and measuring the Hamiltonian. The complexity of such operations can be reduced using two-step low-rank factorizations of the Hamiltonian, accompanied by truncation of small terms [2]. As an application, we will examine the combination [3] of low-rank factorizations with quantum filter diagonalization (QFD). The combination of low-rank factorizations and QDF requires reasonably short circuit depths and modest measurement cost, and can provide accurate predictions for low-lying eigenvalues, when circuit reduction and post-selection error mitigation strategies are deployed.

Another challenge in quantum ES simulations is the presence of dynamical electronic correlation.

As an example of a strategy to treat dynamical electronic correlation on contemporary quantum devices, we will examine the integration of quantum ES algorithms in the workflow of N-electron valence perturbation theory (NEVPT2) [4]. As an application, we will examine the relative stability of OH- and OH. The quantum elevation of NEVPT2 correctly predicts OH- to be more stable than OH, if basis sets with diffuse orbitals are employed.

[1] B. Bauer et al, Chem. Rev. 120, 12685–12717 (2020)

[2] M. Motta et al, npj Quantum Inf. 7, 83 (2021)

[3] J. Cohn, M. Motta and R. Parrish, PRX Quantum 2, 040352 (2021)

[4] A. Tammaro, D. E. Galli, J. Rice and M. Motta, arXiv:2202.13002 (2022)

About the Speaker:

Mario Motta obtained a Ph.D. in theoretical physics in 2015 from the University of Milan (Italy) in the group of professor Davide Galli. Between 2016 and 2019 he was a postdoc in the groups of professors Shiwei Zhang (College of William and Mary) and Garnet Chan (Caltech), focused on classical and quantum computational methods for many-electron systems.

In 2019 he joined IBM research -- Almaden as a research staff member, working on quantum simulation of electronic structure.