{"id":"arxiv/2606.04327","url":"https://arxiv.org/abs/2606.04327","tags":["cs.LG","cs.AI","math.OC"],"score":0,"title":"A Geometric Characterization of the Stationary Plateau for Two-Layer Neural Networks","domain":"cs.LG","source":"arxiv","authors":["Tian Ding","Dawei Li","Ruoyu Sun"],"reviews":0,"subject":"cs.LG","summary":"We investigate the geometric structure of stationary plateaus that arise in the loss landscape of two-layer neural networks with smooth activation functions. We focus on the phenomenon of \"neuron splitting\" where duplicating a hidden neuron yields an affine set of stationary points in a wider network. We provide a comprehensive classification of all stationary points on these plateaus, determining under what conditions they constitute local minima or saddle points. Our characterization hinges on a per-neuron curvature object we term the \"inner Hessian\" matrix. Our analysis reveals that the…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-06-03","reactions":{}}