Overlap-ADAPT-VQE: Practical Quantum Chemistry on Quantum Computers via Overlap-Guided Compact Ansätze - Laboratoire Jacques-Louis Lions
Article Dans Une Revue Communications Physics Année : 2023

Overlap-ADAPT-VQE: Practical Quantum Chemistry on Quantum Computers via Overlap-Guided Compact Ansätze

Résumé

ADAPT-VQE is a robust algorithm for hybrid quantum-classical simulations of quantum chemical systems on near-term quantum computers. While its iterative process systematically reaches the ground state energy, practical implementations of ADAPT-VQE are sensitive to local energy minima, leading to over-parameterized ansätze. We introduce the Overlap-ADAPT-VQE to grow wave-functions by maximizing their overlap with any intermediate target wave-function that already captures some electronic correlation. By avoiding building the ansatz in the energy landscape strewn with local minima, the Overlap-ADAPTVQE produces ultra-compact ansätze suitable for high accuracy initialization of a new ADAPT procedure. Significant advantages over ADAPT-VQE are observed for strongly correlated systems including substantial savings in circuit depth. Since this compression strategy can also be initialized with accurate Selected-Configuration Interaction (SCI) classical target wave-functions, it paves the way for chemically accurate simulations of larger systems, and strengthens the promise of decisively surpassing classical quantum chemistry through the power of quantum computing.
Fichier principal
Vignette du fichier
s42005-023-01312-y.pdf (2.14 Mo) Télécharger le fichier
Origine Fichiers éditeurs autorisés sur une archive ouverte
licence

Dates et versions

hal-03955111 , version 1 (23-09-2024)

Licence

Identifiants

Citer

César Feniou, Muhammad Hassan, Diata Traoré, Emmanuel Giner, Yvon Maday, et al.. Overlap-ADAPT-VQE: Practical Quantum Chemistry on Quantum Computers via Overlap-Guided Compact Ansätze. Communications Physics, 2023, 6 (1), pp.192. ⟨10.1038/s42005-023-01312-y⟩. ⟨hal-03955111⟩
174 Consultations
14 Téléchargements

Altmetric

Partager

More