Rapidity-Coupled Spin Dynamics in Pulsed Laser Fields from Physics-Informed Neural Networks

This paper demonstrates how physics-informed neural networks can solve electron spin dynamics in few-cycle laser pulses. For an electron initially at rest in a CEP-stabilized linearly polarized plane-wave pulse, the rest-frame polarization angle follows the instantaneous vector potential. The analytical solution is derived and validated, showing that net spin rotation vanishes for every carrier-envelope phase value. This work provides a benchmark for spin solvers in ultrafast laser-driven electron source applications.