A Dirac-Frenkel-Onsager principle: Instantaneous residual minimization with gauge momentum for nonlinear parametrizations of PDE solutions
Not on the map. Why?
Abstract (arXiv)
Dirac-Frenkel instantaneous residual minimization evolves nonlinear parametrizations of PDE solutions in time, but ill-conditioning can render the parameter dynamics non-unique. We interpret this non-uniqueness as a gauge freedom: nullspace directions that leave the time derivative unchanged can be used to select better-conditioned parameter velocities. Building on Onsager's minimum-dissipation principle, we introduce a history variable -- interpretable as momentum -- and inject it only along the nullspace directions. The resulting Dirac-Frenkel-Onsager dynamics preserve instantaneous residual minimization, in contrast to standard regularization that can introduce bias, while promoting temporally smooth parameter evolutions. Examples demonstrate that the approach leads to increased robustness in singular and near-singular regimes.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 11 of its 38 references that are in the map. First posted on arXiv 30 Apr 2026.
Builds on
11 of 38 references.
References · 11
- 2017Solving high-dimensional partial differential equations using deep learningcited by 549
- 2022Neural Galerkin Schemes with Active Learning for High-Dimensional Evolution Equationscited by 55
- 2021Evolutional Deep Neural Networkcited by 48
- 2022Implicit Neural Spatial Representations for Time-dependent PDEscited by 35
- 2024A Kaczmarz-inspired approach to accelerate the optimization of neural network wavefunctionscited by 24
- 2023Randomized Sparse Neural Galerkin Schemes for Solving Evolution Equations with Deep Networkscited by 18
- 2023A Stable and Scalable Method for Solving Initial Value PDEs with Neural Networkscited by 15
- 2020Particle-based Energetic Variational Inferencecited by 14
- 2025Improving Energy Natural Gradient Descent through Woodbury, Momentum, and Randomizationcited by 13
- 2023Coupling parameter and particle dynamics for adaptive sampling in Neural Galerkin schemescited by 10
- 2024Sequential-in-time training of nonlinear parametrizations for solving time-dependent partial differential equationscited by 7
Built on by
Citing papers per year: 2026: 1.
1 paper.