activity
20232026
most citedRandomized Sparse Neural Galerkin Schemes for Solving Evolution Equations with Deep Networks

1 citations · 3 across the 12 of their papers we have counts for

collaborators
Showing math.NAShow all

8 papers · 1 filter

math.NA2026

Dirac-Frenkel dynamics with inertia for nonlinearly parametrized solutions of evolution problems

Matteo Raviola, Benjamin Peherstorfer

Even when Dirac-Frenkel dynamics determine a well-defined evolution in function space, the corresponding parameter dynamics can be non-unique or ill-conditioned for redundant nonli…

math.NA2025

Randomized time stepping of nonlinearly parametrized solutions of evolution problems

Yijun Dong, Paul Schwerdtner, Benjamin Peherstorfer

The Dirac-Frenkel variational principle is a widely used building block for using nonlinear parametrizations in the context of model reduction and numerically solving partial diffe…

math.NA2024

Nonlinear model reduction with Neural Galerkin schemes on quadratic manifolds

Philipp Weder, Paul Schwerdtner, Benjamin Peherstorfer

Leveraging nonlinear parametrizations for model reduction can overcome the Kolmogorov barrier that affects transport-dominated problems. In this work, we build on the reduced dynam…

math.NA2024

Empirical sparse regression on quadratic manifolds

Paul Schwerdtner, Serkan Gugercin, Benjamin Peherstorfer

Approximating field variables and data vectors from sparse samples is a key challenge in computational science. Widely used methods such as gappy proper orthogonal decomposition an…

math.NA20241 cited

Online learning of quadratic manifolds from streaming data for nonlinear dimensionality reduction and nonlinear model reduction

Paul Schwerdtner, Prakash Mohan, Aleksandra Pachalieva +3

This work introduces an online greedy method for constructing quadratic manifolds from streaming data, designed to enable in-situ analysis of numerical simulation data on the Petab…

math.NA2024

Sequential-in-time training of nonlinear parametrizations for solving time-dependent partial differential equations

Huan Zhang, Yifan Chen, Eric Vanden-Eijnden +1

Sequential-in-time methods solve a sequence of training problems to fit nonlinear parametrizations such as neural networks to approximate solution trajectories of partial different…