Accelerating wavepacket propagation with machine learning

verfasst von

Kanishka Singh, Ka Hei Lee, Daniel Peláez, Annika Bande

Abstract

In this work, we discuss the use of a recently introduced machine learning (ML) technique known as Fourier neural operators (FNO) as an efficient alternative to the traditional solution of the time-dependent Schrödinger equation (TDSE). FNOs are ML models which are employed in the approximated solution of partial differential equations. For a wavepacket propagating in an anharmonic potential and for a tunneling system, we show that the FNO approach can accurately and faithfully model wavepacket propagation via the density. Additionally, we demonstrate that FNOs can be a suitable replacement for traditional TDSE solvers in cases where the results of the quantum dynamical simulation are required repeatedly such as in the case of parameter optimization problems (e.g., control). The speed-up from the FNO method allows for its combination with the Markov-chain Monte Carlo approach in applications that involve solving inverse problems such as optimal and coherent laser control of the outcome of dynamical processes.

Organisationseinheit(en)

Institut für Anorganische Chemie
PhoenixD: Simulation, Fabrikation und Anwendung optischer Systeme

Externe Organisation(en)

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH
Freie Universität Berlin (FU Berlin)
Universität Paris-Süd

Typ

Artikel

Journal

Journal of computational chemistry

Band

45

Seiten

2360-2373

Anzahl der Seiten

14

ISSN

0192-8651

Publikationsdatum

02.09.2024

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

ASJC Scopus Sachgebiete

Chemie (insg.), Computational Mathematics

Elektronische Version(en)

https://doi.org/10.1002/jcc.27443 (Zugang: Geschlossen)