Metadata only
Datum
2023-01Typ
- Report
ETH Bibliographie
yes
Altmetrics
Abstract
A large class of inverse problems for PDEs are only well-defined as mappings from operators to functions. Existing operator learning frameworks map functions to functions and need to be modified to learn inverse maps from data. We propose a novel architecture termed Neural Inverse Operators (NIOs) to solve these PDE inverse problems. Motivated by the underlying mathematical structure, NIO is based on a suitable composition of DeepONets and FNOs to approximate mappings from operators to functions. A variety of experiments are presented to demonstrate that NIOs significantly outperform baselines and solve PDE inverse problems robustly, accurately and are several orders of magnitude faster than existing direct and PDE-constrained optimization methods. Mehr anzeigen
Publikationsstatus
publishedExterne Links
Zeitschrift / Serie
SAM Research ReportBand
Verlag
Seminar for Applied Mathematics, ETH ZurichOrganisationseinheit
03851 - Mishra, Siddhartha / Mishra, Siddhartha
02889 - ETH Institut für Theoretische Studien / ETH Institute for Theoretical Studies
Zugehörige Publikationen und Daten
Is previous version of: http://hdl.handle.net/20.500.11850/656484
ETH Bibliographie
yes
Altmetrics