Physics-informed neural networks for solving saddle-point equations in strong-field physics with tailored fields

This work develops an unsupervised physics-informed neural-network framework for solving complex-time saddle-point equations governing direct above-threshold ionisation in tailored laser fields. A window parametrisation strategy guides optimisation towards physically relevant saddle points and improves convergence stability.

DOI · arXiv

Recommended citation: Jiakang Chen, Sufia Hashim, and Carla Figueira de Morisson Faria. (2026). "Physics-informed neural networks for solving saddle-point equations in strong-field physics with tailored fields." Physical Review Research (accepted).
Download Paper