dadesso17/RJ-PINNs

RJ-PINNs, introduced by Dadoyi Dadesso, is a pioneering framework using Jacobian-based least-squares (TRF) to directly minimize residuals without traditional loss functions. This method is the first to eliminate gradient optimizers in PINNs, offering unmatched robustness for inverse PDE problems.

Language: Python - Size: 476 KB - Last synced at: 25 days ago - Pushed at: 25 days ago - Stars: 2 - Forks: 0