Copertina

DyKAF: Dynamical Kronecker Approximation of the Fisher Information Matrix for Gradient Preconditioning

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Yudin, Nikolay, Grishina, Ekaterina, Veprikov, Andrey, Beznosikov, Alexandr, Rakhuba, Maxim
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning Numerical Analysis Optimization and Control 68T07, 65F55, 15A18
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

COALA: Numerically Stable and Efficient Framework for Context-Aware Low-Rank Approximation
di: Parkina, Uliana, et al.
Pubblicazione: (2025)

Multiscale Neural Networks for Approximating Green's Functions
di: Hao, Wenrui, et al.
Pubblicazione: (2024)

Accelerating Newton-Schulz Iteration for Orthogonalization via Chebyshev-type Polynomials
di: Grishina, Ekaterina, et al.
Pubblicazione: (2025)

Expansive Natural Neural Gradient Flows for Energy Minimization
di: Dahmen, Wolfgang, et al.
Pubblicazione: (2025)

A discontinuous Galerkin plane wave neural network method for Helmholtz equation and Maxwell's equations
di: Yuan, Long, et al.
Pubblicazione: (2025)

A Neural Multigrid Solver for Helmholtz Equations with High Wavenumber and Heterogeneous Media
di: Cui, Chen, et al.
Pubblicazione: (2024)

Discontinuous hybrid neural networks for the one-dimensional partial differential equations
di: Wang, Xiaoyu, et al.
Pubblicazione: (2025)

Space-time parallel scaling of Parareal with a physics-informed Fourier Neural Operator coarse propagator applied to the Black-Scholes equation
di: Ibrahim, Abdul Qadir, et al.
Pubblicazione: (2024)

Learning a generalized multiscale prolongation operator
di: Liu, Yucheng, et al.
Pubblicazione: (2024)

A framework of discontinuous Galerkin neural networks for iteratively approximating residuals
di: Yuan, Long, et al.
Pubblicazione: (2025)

Leveraging Operator Learning to Accelerate Convergence of the Preconditioned Conjugate Gradient Method
di: Kopaničáková, Alena, et al.
Pubblicazione: (2025)

LoRA meets Riemannion: Muon Optimizer for Parametrization-independent Low-Rank Adapters
di: Bogachev, Vladimir, et al.
Pubblicazione: (2025)

Coupling-Robust Accuracy in Multiphysics Physics Informed Neural Networks via Kronecker-Preconditioned Optimization
di: Park, Youngjae, et al.
Pubblicazione: (2026)

Neural Approximate Inverse Preconditioners
di: Xu, Tianshi, et al.
Pubblicazione: (2025)

SVD-Preconditioned Gradient Descent Method for Solving Nonlinear Least Squares Problems
di: Chang, Zhipeng, et al.
Pubblicazione: (2026)

Pay Attention to Attention Distribution: A New Local Lipschitz Bound for Transformers
di: Yudin, Nikolay, et al.
Pubblicazione: (2025)

A Neural-Operator Preconditioned Newton Method for Accelerated Nonlinear Solvers
di: Lee, Youngkyu, et al.
Pubblicazione: (2025)

Randomized Matrix Sketching for Neural Network Training and Gradient Monitoring
di: Antil, Harbir, et al.
Pubblicazione: (2025)

Multifidelity domain decomposition-based physics-informed neural networks and operators for time-dependent problems
di: Heinlein, Alexander, et al.
Pubblicazione: (2024)

A Paired Autoencoder Framework for Inverse Problems via Bayes Risk Minimization
di: Hart, Emma, et al.
Pubblicazione: (2025)

Towards Model Discovery Using Domain Decomposition and PINNs
di: Saha, Tirtho S., et al.
Pubblicazione: (2024)

A neuron-wise subspace correction method for the finite neuron method
di: Park, Jongho, et al.
Pubblicazione: (2022)

Neural Preconditioning via Krylov Subspace Geometry
di: Dimola, Nunzio, et al.
Pubblicazione: (2025)

A Flow-rate-conserving CNN-based Domain Decomposition Method for Blood Flow Simulations
di: Klaes, Simon, et al.
Pubblicazione: (2025)

Adaptive Multilevel Neural Networks for Parametric PDEs with Error Estimation
di: Schütte, Janina E., et al.
Pubblicazione: (2024)

Accuracy and stability of Artificial Neural Networks for HP-Splines frequency parameter selection
di: Bruni, Vittoria, et al.
Pubblicazione: (2026)

A reduced order Schwarz method for nonlinear multiscale elliptic equations based on two-layer neural networks
di: Chen, Shi, et al.
Pubblicazione: (2021)

Non-Asymptotic Analysis of Projected Gradient Descent for Physics-Informed Neural Networks
di: Nießen, Jonas, et al.
Pubblicazione: (2025)

The Neural Approximated Virtual Element Method for Elasticity Problems
di: Berrone, Stefano, et al.
Pubblicazione: (2025)

Two-level overlapping additive Schwarz preconditioner for training scientific machine learning applications
di: Lee, Youngkyu, et al.
Pubblicazione: (2024)

Enhancing training of physics-informed neural networks using domain-decomposition based preconditioning strategies
di: Kopaničáková, Alena, et al.
Pubblicazione: (2023)

Diffusion-based Models for Unpaired Super-resolution in Fluid Dynamics
di: Xu, Wuzhe, et al.
Pubblicazione: (2025)

An Additively Preconditioned Trust Region Strategy for Machine Learning
di: Alegría, Samuel Cruz, et al.
Pubblicazione: (2025)

Multilevel Training for Kolmogorov Arnold Networks
di: Southworth, Ben S., et al.
Pubblicazione: (2026)

Deep Micro Solvers for Rough-Wall Stokes Flow in a Heterogeneous Multiscale Method
di: Ström, Emanuel, et al.
Pubblicazione: (2025)

Improving Sketching Algorithms for Low-Rank Matrix Approximation via Sketch-Power Iterations
di: Chang, Chao, et al.
Pubblicazione: (2026)

Randomized Large-Scale Quaternion Matrix Approximation: Practical Rangefinders and One-Pass Algorithm
di: Chang, Chao, et al.
Pubblicazione: (2024)

Model Parallel Training and Transfer Learning for Convolutional Neural Networks by Domain Decomposition
di: Klawonn, Axel, et al.
Pubblicazione: (2024)

Domain-decomposed image classification algorithms using linear discriminant analysis and convolutional neural networks
di: Klawonn, Axel, et al.
Pubblicazione: (2024)

Objective Value Change and Shape-Based Accelerated Optimization for the Neural Network Approximation
di: Xie, Pengcheng, et al.
Pubblicazione: (2025)