{"id":"arxiv/2609.26534","url":"https://arxiv.org/abs/2609.26534","tags":["math.NA"],"score":0,"title":"Error Analysis for Solving Elliptic Interface Problems with Randomized Neural Networks","domain":"math.NA","source":"arxiv","authors":["Jun Hu","Sidi Wu"],"reviews":0,"subject":"math.NA","summary":"This paper presents a unified error analysis of randomized neural networks (RaNNs) for elliptic interface problems. We first derive an integral representation of Barron functions using the standard $\\tanh$ activation, which yields a $C^2$-approximation bound valid for any hidden-parameter distribution supported on a cube with strictly positive density. We then establish a quantitative $\\mathcal{O}(n^{-1})$ generalization bound for the composite loss that enforces the partial differential equation, boundary, and interface conditions simultaneously, where $n$ denotes the number of training…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-09-22","reactions":{}}