Christopher K. Long, Kieran Dalton, Yordan S. Yordanov, Charles G. Smith, Nicholas J. Mayhall, Sophia E. Economou, Edwin Barnes, Frederico Martins, Crispin H. W. Barnes, David R. M. Arvidsson-Shukur, and Normann Mertig
Date & time: 25 February 2025 10:30–11:00 GMT
Location: University of Cambridge, Cambridge, United Kingdom
Conference: Hitachi Cambridge Quantum Workshop
I presented work from Refs. [1–3].
Download slide deck: PPTX1, PDF1
Abstract
Noisy intermediate-scale quantum (NISQ) devices are hoped to yield preliminary advantages over classical computers before fault-tolerant quantum devices emerge. Variational approaches largely dominate the space of NISQ-tailored algorithms due to their short coherence time requirements and robustness to systematic errors. Throughout this talk, we will focus on the application of variational quantum algorithms to quantum computational chemistry. The recent theoretical advances in NISQ algorithms have been complimented (sic) by the rapid development of quantum processors utilising a variety of physical systems. Silicon quantum processors constitute a particularly promising platform due to high levels of tunability and the potential for rapid industrial scaling. In this talk, I will present numerical evidence suggesting that current variational quantum algorithms for quantum computational chemistry are not viable in the presence of realistic noise levels [1]. There are three possible directions for improvement: First, I will introduce error mitigation [1]. Second, I will present a general framework for designing noise-robust variational quantum algorithms [2]. We will see that neither error mitigation nor algorithmic improvements are sufficient to perform useful chemical computations. Third, I will relax the digital gate-based model of quantum computation to control at the level of the device Hamiltonian (pulse-based computation) [3–5].2 We will find silicon quantum processors boast extremely competitive state preparation times for chemical ground states [3] when compared with superconducting hardware [4,5].2 This dramatic reduction in state-preparation time from the digital gate-based model increases the coherence time budget for computations and makes significant steps towards chemical simulations on quantum processors.
Other versions
- Towards chemistry simulations on silicon quantum processors (Hersonissos)
- Towards chemistry simulations on silicon quantum processors (UCL)