Imagen de Portada

Geometrical structures of digital fluctuations in parameter space of neural networks trained with adaptive momentum optimization

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autor principal:	Netay, Igor V.
Formato:	Preprint
Publicado:	2024
Materias:	Machine Learning Numerical Analysis
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Algorithms and data structures for automatic precision estimation of neural networks
por: Netay, Igor V.
Publicado: (2025)

Algorithms and data structures for numerical computations with automatic precision estimation
por: Netay, Igor V.
Publicado: (2024)

Gradient-free training of recurrent neural networks
por: Bolager, Erik Lien, et al.
Publicado: (2024)

DPG loss functions for learning parameter-to-solution maps by neural networks
por: Castillo, Pablo Cortés, et al.
Publicado: (2025)

Domain decomposition architectures and Gauss-Newton training for physics-informed neural networks
por: Heinlein, Alexander, et al.
Publicado: (2025)

Averaged Adam accelerates stochastic optimization in the training of deep neural network approximations for partial differential equation and optimal control problems
por: Dereich, Steffen, et al.
Publicado: (2025)

On the optimal approximation of Sobolev and Besov functions using deep ReLU neural networks
por: Yang, Yunfei
Publicado: (2024)

An extended physics informed neural network for preliminary analysis of parametric optimal control problems
por: Demo, Nicola, et al.
Publicado: (2021)

Fixed-budget online adaptive learning for physics-informed neural networks. Towards parameterized problem inference
por: Nguyen, Thi Nguyen Khoa, et al.
Publicado: (2022)

A decomposition-based robust training of physics-informed neural networks for nearly incompressible linear elasticity
por: Dick, Josef, et al.
Publicado: (2025)

Stochastic diagonal estimation with adaptive parameter selection
por: Han, Zongyuan, et al.
Publicado: (2024)

Convergence rates for gradient descent in the training of overparameterized artificial neural networks with piecewise affine activation
por: Jentzen, Arnulf, et al.
Publicado: (2021)

VS-PINN: A fast and efficient training of physics-informed neural networks using variable-scaling methods for solving PDEs with stiff behavior
por: Ko, Seungchan, et al.
Publicado: (2024)

Generalizing the SINDy approach with nested neural networks
por: Fiorini, Camilla, et al.
Publicado: (2024)

Training Hamiltonian neural networks without backpropagation
por: Rahma, Atamert, et al.
Publicado: (2024)

Series of quasi-uniform scatterings with fast search, root systems and neural network classifications
por: Netay, Igor V.
Publicado: (2025)

Stable neural networks and connections to continuous dynamical systems
por: Ehrhardt, Matthias J., et al.
Publicado: (2025)

Energy stable neural network for gradient flow equations
por: Wu, Yue, et al.
Publicado: (2023)

Pseudo-differential-enhanced physics-informed neural networks
por: Gracyk, Andrew
Publicado: (2026)

A fast neural hybrid Newton solver adapted to implicit methods for nonlinear dynamics
por: Jin, Tianyu, et al.
Publicado: (2024)

Group-invariant tensor train networks for supervised learning
por: Sprangers, Brent, et al.
Publicado: (2022)

Consistent machine learning for topology optimization with microstructure-dependent neural network material models
por: Vijayakumaran, Harikrishnan, et al.
Publicado: (2024)

Exact and approximate error bounds for physics-informed neural networks
por: Chantada, Augusto T., et al.
Publicado: (2024)

Pseudo-Hamiltonian neural networks for learning partial differential equations
por: Eidnes, Sølve, et al.
Publicado: (2023)

Physics-informed neural networks for operator equations with stochastic data
por: Escapil-Inchauspé, Paul, et al.
Publicado: (2022)

Tensor train completion: local recovery guarantees via Riemannian optimization
por: Budzinskiy, Stanislav, et al.
Publicado: (2021)

Sequential decoder training for improved latent space dynamics identification
por: Anderson, William, et al.
Publicado: (2025)

Gradient-free neural topology optimization: Towards effective fracture-resistant designs
por: Kus, Gawel, et al.
Publicado: (2024)

Astral: training physics-informed neural networks with error majorants
por: Fanaskov, Vladimir, et al.
Publicado: (2024)

A note on the adjoint method for neural ordinary differential equation network
por: Hu, Pipi
Publicado: (2024)

Two-hidden-layer ReLU neural networks and finite elements
por: Jin, Pengzhan
Publicado: (2024)

Solving engineering eigenvalue problems with neural networks using the Rayleigh quotient
por: Rowan, Conor, et al.
Publicado: (2025)

Orthogonal greedy algorithm for linear operator learning with shallow neural network
por: Lin, Ye, et al.
Publicado: (2025)

Macroscopic auxiliary asymptotic preserving neural networks for the linear radiative transfer equations
por: Li, Hongyan, et al.
Publicado: (2024)

Memorization capacity of deep ReLU neural networks characterized by width and depth
por: Yang, Xin, et al.
Publicado: (2026)

A local approach to parameter space reduction for regression and classification tasks
por: Romor, Francesco, et al.
Publicado: (2021)

Variational operator learning: A unified paradigm marrying training neural operators and solving partial differential equations
por: Xu, Tengfei, et al.
Publicado: (2023)

Towards optimal hierarchical training of neural networks
por: Feischl, Michael, et al.
Publicado: (2024)

CoLoRA: Continuous low-rank adaptation for reduced implicit neural modeling of parameterized partial differential equations
por: Berman, Jules, et al.
Publicado: (2024)

Structure-preserving neural networks for the regularized entropy-based closure of the Boltzmann moment system
por: Schotthöfer, Steffen, et al.
Publicado: (2024)