{"id":"arxiv/2603.21568","url":"https://arxiv.org/abs/2603.21568","tags":["math.NA","cs.LG","math.DS"],"score":0,"title":"Stability and Bifurcation Analysis of Nonlinear PDEs via Random Projection-based PINNs: A Krylov-Arnoldi Approach","domain":"math.NA","source":"arxiv","authors":["Gianluca Fabiani","Michail E. Kavousanakis","Constantinos Siettos","Ioannis G. Kevrekidis"],"reviews":0,"subject":"math.NA","summary":"We present a numerical framework for the stability and bifurcation analysis of nonlinear partial differential equations (PDEs) in which the solutions are sought in the function space spanned by physics-informed random projection neural networks (PI-RPNNs), and discretized via a collocation approach. These are shallow, single-hidden-layer networks with randomly sampled and fixed a priori hidden-layer weights; only the linear output layer weights are optimized, reducing training to a single least-squares solve. This linear output structure enables the explicit generalized eigenvalue…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-23","reactions":{}}