KIT | KIT-Bibliothek | Impressum | Datenschutz

Generalization properties of neural network approximations to frustrated magnet ground states

Westerhout, T.; Astrakhantsev, N.; Tikhonov, K. S.; Katsnelson, M. I.; Bagrov, A. A.

Abstract:
Neural quantum states (NQS) attract a lot of attention due to their potential to serve as a very expressive variational ansatz for quantum many-body systems. Here we study the main factors governing the applicability of NQS to frustrated magnets by training neural networks to approximate ground states of several moderately-sized Hamiltonians using the corresponding wave function structure on a small subset of the Hilbert space basis as training dataset. We notice that generalization quality, i.e. the ability to learn from a limited number of samples and correctly approximate the target state on the rest of the space, drops abruptly when frustration is increased. We also show that learning the sign structure is considerably more difficult than learning amplitudes. Finally, we conclude that the main issue to be addressed at this stage, in order to use the method of NQS for simulating realistic models, is that of generalization rather than expressibility.

Open Access Logo


Verlagsausgabe §
DOI: 10.5445/IR/1000118190
Veröffentlicht am 21.04.2020
Originalveröffentlichung
DOI: 10.1038/s41467-020-15402-w
Web of Science
Zitationen: 2
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Nanotechnologie (INT)
Publikationstyp Zeitschriftenaufsatz
Publikationsjahr 2020
Sprache Englisch
Identifikator ISSN: 2041-1723
KITopen-ID: 1000118190
HGF-Programm 43.21.02 (POF III, LK 01)
Quantum Properties of Nanostructures
Erschienen in Nature Communications
Band 11
Heft 1
Seiten Artikel-Nr.: 1593
Vorab online veröffentlicht am 27.03.2020
Nachgewiesen in Web of Science
Scopus
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page