{"id":"arxiv/2603.16746","url":"https://arxiv.org/abs/2603.16746","tags":["math.DS","cs.LG","nlin.CD"],"score":0,"title":"Data-driven forced response analysis with min-max representations of nonlinear restoring forces","domain":"math.DS","source":"arxiv","authors":["Akira Saito","Hiromu Fujita"],"reviews":0,"subject":"math.DS","summary":"This paper discusses a novel data-driven nonlinearity identification method for mechanical systems with nonlinear restoring forces such as polynomial, piecewise-linear, and general displacement-dependent nonlinearities. The proposed method is built upon the universal approximation theorem that states that a nonlinear function can be approximated by a linear combination of activation functions in artificial neural network framework. The proposed approach utilizes piecewise linear springs with initial gaps to act as the activation functions of the neurons of artificial neural networks. A…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-17","reactions":{}}