Regensburg 2025 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

MM: Fachverband Metall- und Materialphysik

MM 9: Poster

MM 9.61: Poster

Montag, 17. März 2025, 18:30–20:30, P1

Learning the Reduced Density Matrix Functional from Quantum Processors and Using Density Matrix Embedding Theory to Extend its Universality — •Martin Uttendorfer — Deutsches Zentrum für Luft- und Raumfahrt (DLR), Köln, Deutschland

The advent of quantum computing makes reduced density matrix functional theory (RDMFT) on an exact level viable. By directly applying RDMFT, derived from Levy-Lieb's constrained search, this work leverages quantum processors to overcome some of DFT*s shortcomings, enabling more accurate modeling of quantum chemical and condensed matter systems. The proposed approach incorporates quantum algorithms that utilize variational quantum eigensolvers (VQE), which is viable to be executed on near-term intermediate scale quantum devices (NISQ) in conjunction with machine learning techniques. This work examines the theory*s application to different particle types, including fermions, bosons, and hard-core bosons, highlighting the flexibility of the RDMFT framework. Additionally, density matrix embedding theory (DMET) is incorporated, allowing for a hybrid classical-quantum approach that extends the functional*s universality. This work presents a quantum algorithmic approach to obtain the functional and provides a computational strategy for studying complex many-body systems while keeping the use of limited quantum resources to a minimum.

Keywords: Quantum Computing; Density Functional Theory; Reduced Density Matrix Functional Theory; Density Matrix Embedding Theory