Open access
Nov 2026
A two-step physics-informed neural network approach for open-loop optimal control of the Gardner equation with limited data and partially unknown parameters
This paper introduces a two-stage PINN framework, using a Lagrangian multiplier framework, and employs PINNs to jointly approximate the state, adjoint, and control variables, enabling effective learning even under severe data scarcity.
Sana Idrees, Shahbaz Ahmad
· Machine Learning for Computa... · 0 citations