Neural Operators for Elliptic PDEs in Polytopes and Operator Networks via Fixed Point Iterations
27 Mar 2026
03.30 PM - 04.30 PM
SPMS-LT5 (SPMS-03-08)
Current Students
Abstract
We review recent results and ongoing research on numerical analysis of neural operators, in the references below.
References:
Neural General Operator Networks via Banach Fixed Point Iterations M Feischl, Ch Schwab, F Zehetgruber Research Report 2025-13, SAM, ETH Zurich, April 2025 (to appear 2026)
Neural and spectral operator surrogates: construction and expression rate bounds L Herrmann, Ch Schwab, J Zech Advances in Computational Mathematics 50 (4), 72, 2024
Exponential convergence of deep operator networks for elliptic partial differential equations C Marcati, Ch Schwab SIAM Journal on Numerical Analysis 61 (3), 1513-1545, 2024
Expression Rates of Neural Operators for Linear Elliptic PDEs in Polytopes C Marcati, C Schwab arXiv preprint arXiv:2409.17552 =
Biography
Professor Christoph Schwab is a leading expert in numerical analysis and scientific computing. His research interests include space-time compressive discretisations of evolution equations, PDEs with random input data, high-dimensional numerics for PDEs with multiple scales, sparse tensor approximations of high-dimensional and stochastic PDEs, multilevel Monte Carlo and Quasi-Monte Carlo algorithms for PDEs, deep neural network approximation.