Imagen de Portada

A Fourier perspective on the learning dynamics of neural networks: from sample complexities to mechanistic insights

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Ricci, Fabiola, Merger, Claudia, Goldt, Sebastian
Formato:	Preprint
Publicado:	2026
Materias:	Machine Learning Disordered Systems and Neural Networks Statistical Mechanics Probability
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Feature learning from non-Gaussian inputs: the case of Independent Component Analysis in high dimensions
por: Ricci, Fabiola, et al.
Publicado: (2025)

Factual recall in linear associative memories: sharp asymptotics and mechanistic insights
por: Giorlandino, Alessio, et al.
Publicado: (2026)

Generalization Dynamics of Linear Diffusion Models
por: Merger, Claudia, et al.
Publicado: (2025)

Two failure modes of deep transformers and how to avoid them: a unified theory of signal propagation at initialisation
por: Giorlandino, Alessio, et al.
Publicado: (2025)

Memorisation, convergence and generalisation in generative models
por: Maillard, Antoine, et al.
Publicado: (2026)

Sampling the space of solutions of an artificial neural network
por: Zambon, Alessandro, et al.
Publicado: (2025)

Coding schemes in neural networks learning classification tasks
por: van Meegen, Alexander, et al.
Publicado: (2024)

Sampling with flows, diffusion and autoregressive neural networks: A spin-glass perspective
por: Ghio, Davide, et al.
Publicado: (2023)

Spring-block theory of feature learning in deep neural networks
por: Shi, Cheng, et al.
Publicado: (2024)

Mapping of attention mechanisms to a generalized Potts model
por: Rende, Riccardo, et al.
Publicado: (2023)

The twin peaks of learning neural networks
por: Demyanenko, Elizaveta, et al.
Publicado: (2024)

High-dimensional manifold of solutions in neural networks: insights from statistical physics
por: Malatesta, Enrico M.
Publicado: (2023)

Phase transitions in the mini-batch size for sparse and dense two-layer neural networks
por: Marino, Raffaele, et al.
Publicado: (2023)

Diffusive geodesics wandering in networks of rigid chains
por: Marquis, Ulysse
Publicado: (2025)

A spin-glass model for the loss surfaces of generative adversarial networks
por: Baskerville, Nicholas P, et al.
Publicado: (2021)

On the role of non-linear latent features in bipartite generative neural networks
por: Bonnaire, Tony, et al.
Publicado: (2025)

Stochastic Gradient Descent-like relaxation is equivalent to Metropolis dynamics in discrete optimization and inference problems
por: Angelini, Maria Chiara, et al.
Publicado: (2023)

Gaussian mixtures and non-parametric likelihoods through the lens of statistical mechanics
por: Ghosh, Subhroshekhar, et al.
Publicado: (2026)

The autoregressive neural network architecture of the Boltzmann distribution of pairwise interacting spins systems
por: Biazzo, Indaco
Publicado: (2023)

Dynamic neuron approach to deep neural networks: Decoupling neurons for renormalization group analysis
por: Lee, Donghee, et al.
Publicado: (2024)

Weight fluctuations in (deep) linear neural networks and a derivation of the inverse-variance flatness relation
por: Gross, Markus, et al.
Publicado: (2023)

Escape dynamics and implicit bias of one-pass SGD in overparameterized quadratic networks
por: Bocchi, Dario, et al.
Publicado: (2026)

Microscopic and collective signatures of feature learning in neural networks
por: Corti, Andrea, et al.
Publicado: (2025)

Drift-Diffusion Matching: Embedding dynamics in latent manifolds of asymmetric neural networks
por: Nartallo-Kaluarachchi, Ramón, et al.
Publicado: (2026)

Optimal generalisation and learning transition in extensive-width shallow neural networks near interpolation
por: Barbier, Jean, et al.
Publicado: (2025)

The committee machine: Computational to statistical gaps in learning a two-layers neural network
por: Aubin, Benjamin, et al.
Publicado: (2018)

Activation Functions, Statistics and Learning of Higher-Order Interactions in Restricted Boltzmann Machines
por: di Sarra, Giovanni, et al.
Publicado: (2026)

Phase Transitions between Accuracy Regimes in L2 regularized Deep Neural Networks
por: Ersoy, Ibrahim Talha, et al.
Publicado: (2025)

Evolutionary chemical learning in dimerization networks
por: Tkachenko, Alexei V., et al.
Publicado: (2025)

Sharp feature-learning transitions and Bayes-optimal neural scaling laws in extensive-width networks
por: Nguyen, Minh-Toan, et al.
Publicado: (2026)

Introduction to dynamical mean-field theory of randomly connected neural networks with bidirectionally correlated couplings
por: Zou, Wenxuan, et al.
Publicado: (2023)

Correlation thresholds in the steady states of particle systems and spin glasses
por: Calvert, Jacob, et al.
Publicado: (2025)

Structure of solutions to continuous constraint satisfaction problems through the statistics of wedged and inscribed spheres
por: Kent-Dobias, Jaron
Publicado: (2025)

Variational representation and estimates for the free energy of a quenched charged polymer model
por: Poisat, Julien
Publicado: (2025)

Dynamic Space Packing
por: Dandekar, Rahul, et al.
Publicado: (2023)

Dynamic Space Filling
por: Krapivsky, P. L.
Publicado: (2025)

Tensor-Network Population Annealing
por: Oshima, Takumi, et al.
Publicado: (2026)

Deep Neural Nets as Hamiltonians
por: Winer, Mike, et al.
Publicado: (2025)

How does Chain of Thought decompose complex tasks?
por: Nadgir, Amrut, et al.
Publicado: (2026)

Gaussian approximation of dynamic cavity equations for linearly-coupled stochastic dynamics
por: Tarabolo, Mattia, et al.
Publicado: (2024)