Jul 2026
LIGO-PINN: Learned Initialization via Gated Optimization to Alleviate Convergence Failures in Physics Informed Neural Networks
This work proposes a framework for Learned Initialization via Gated Layerwise Optimization (LIGO-PINN) to overcome PINN convergence failures and demonstrates that this methodology outperforms state-of-the-art methods designed to alleviate PINN failures.
N. P. Anurag, S. Adhikari, Taniya Kapoor et al.
· arXiv.org · 0 citations