works on

From the 1 of 18 linked papers with an AI index.

collaborators
Showing math.NAShow all

18 papers · 1 filter

math.NA2026

Advanced Linear Algebra with Applications - Part I (Numerical linear algebra for PDEs, machine learning, and data assimilation)

Victorita Dolean, Jemima Tabeart

These lecture notes form the first part of a master's-level course on advanced numerical linear algebra. Their aim is not only to present the classical algorithms, but to show why…

math.NA2026

When can a neural operator replace a coarse solve? Architectural principles for two-level preconditioning

Hugo Melchers, Victorita Dolean, Michael Abdelmalik

The paper investigates which neural operator architectures can replace the coarse solve in two‑level preconditioners for discretised linear PDEs, finding that the Neural Green's Op…

math.NA2026

Convergence analysis of GMRES applied to Helmholtz problems near resonances

Victorita Dolean, Pierre Marchand, Axel Modave +1

The finite element solution of Helmholtz problems near resonant or quasi-resonant frequencies poses significant challenges, as iterative solvers typically suffer from severely degr…

math.NA2026

Are Deep Learning Based Hybrid PDE Solvers Reliable? Why Training Paradigms and Update Strategies Matter

Yuhan Wu, Jan Willem van Beek, Victorita Dolean +1

Deep learning-based hybrid iterative methods (DL-HIMs) integrate classical numerical solvers with neural operators, utilizing their complementary spectral biases to accelerate conv…

math.NA2026

Spectral coarse spaces based on indefinite operators: the -GenEO method

Théophile Chaumont-Frelet, Victorita Dolean, Mark Fry +2

GenEO (`Generalised Eigenvalue problems on the Overlap') is a method for constructing coarse spaces used in the preconditioning of iterative solvers for discrete PDEs. This method…

math.NA2026

A Guided Tour of Modern Domain Decomposition: From Schwarz Iterations to Robust Preconditioners and HPC Implementations

Victorita Dolean, Pierre Jolivet, Frédéric Nataf +1

Domain decomposition methods (DDMs) provide a unifying framework for the scalable numerical solution of partial differential equations. Originating from Schwarz's alternating metho…