Deutsch
 
Hilfe Datenschutzhinweis Impressum
  DetailsucheBrowse
  1. Datensatz ansehen / 
  2. Datensatz Übersicht

Datensatz

DATENSATZ AKTIONENEXPORT
Zur Ablage hinzufügen
  DetailsÜbersicht

Freigegeben
Preprint

Variational Neural and Tensor Network Approximations of Thermal States

MPG-Autoren
/persons/resource/persons251442

Lu,  Sirui
Theory, Max Planck Institute of Quantum Optics, Max Planck Society;
MCQST - Munich Center for Quantum Science and Technology, External Organizations;
IMPRS (International Max Planck Research School), Max Planck Institute of Quantum Optics, Max Planck Society;

/persons/resource/persons232821

Giudice,  Giacomo
Theory, Max Planck Institute of Quantum Optics, Max Planck Society;
MCQST - Munich Center for Quantum Science and Technology, External Organizations;

/persons/resource/persons60441

Cirac,  J. Ignacio       
Theory, Max Planck Institute of Quantum Optics, Max Planck Society;
MCQST - Munich Center for Quantum Science and Technology, External Organizations;

Externe Ressourcen
Es sind keine externen Ressourcen hinterlegt
Volltexte (beschränkter Zugriff)
Für Ihren IP-Bereich sind aktuell keine Volltexte freigegeben.
Volltexte (frei zugänglich)

2401.14243.pdf
(Preprint), 4MB

Ergänzendes Material (frei zugänglich)
Es sind keine frei zugänglichen Ergänzenden Materialien verfügbar
Zitation

Lu, S., Giudice, G., & Cirac, J. I. (submitted). Variational Neural and Tensor Network Approximations of Thermal States.


Zitierlink: https://hdl.handle.net/21.11116/0000-000E-A830-E
Zusammenfassung
We introduce a variational Monte Carlo algorithm for approximating
finite-temperature quantum many-body systems, based on the minimization of a
modified free energy. We employ a variety of trial states -- both tensor
networks as well as neural networks -- as variational ans\"atze for our
numerical optimization. We benchmark and compare different constructions in the
above classes, both for one- and two-dimensional problems, with systems made of
up to \(N=100\) spins. Despite excellent results in one dimension, our results
suggest that the numerical ans\"atze employed have certain expressive
limitations for tackling more challenging two-dimensional systems.