Physics Informed Neural Networks based Solvers for the Time-Dependent Schrödinger Equation

Physics Informed Neural Networks based Solvers for the Time-Dependent Schrödinger Equation

Shah, K.; Cangi, A.

We demonstrate the utility of Physics Informed Neural Network based solvers for the solution of the Time-Dependent Schrödinger Equation. We study the performance and generalisability of PINN solvers on a simple quantum system. The method developed here can be potentially extended as a surrogate model for Time-Dependent Density Functional Theory, enabling the simulation of large-scale calculations of electron dynamics in matter exposed to strong electromagnetic fields, high temperatures, and pressures.

Keywords: time-dependent density functional theory; physics informed neural networks

  • Poster
    DFT Methods for Matter under Extreme Conditions, 21.02.2022, Görlitz, Germany
  • Poster
    Strongly Coupled Coulomb Systems, 25.07.2022, Görlitz, Germany
  • Poster
    Big data analytical methods for complex systems, 06.10.2022, Wrocław, Poland
  • Poster
    HZDR DocSeminar, 20.10.2022, Wrocław, Poland
  • Poster
    DPG-Frühjahrstagung Sektion Kondensierte Materie, 27.03.2023, Dresden, Germany
  • Poster
    2023 Time Dependent Density Functional Theory School & Workshop: Excited states and dynamics, 29.06.-08.07.2023, Newark, United States of America

Permalink: https://www.hzdr.de/publications/Publ-35349