works on
convergence analysis 1gradient flow training 1inverse problems 1parametric problems 1residual neural networks 1

From the 1 of 3 linked papers with an AI index.

collaborators
Showing math.NAShow all

1 paper · 1 filter

math.NA2026

Approximation of solutions of parameter-dependent problems by residual neural networks

Ana Carpio

The paper introduces a simple training scheme for residual neural networks based on analytic activation functions and gradient flows, with convergence guaranteed by Lojasiewicz the…