Image de couverture de livre

Spectral alignment of stochastic gradient descent for high-dimensional classification tasks

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Arous, Gerard Ben, Gheissari, Reza, Huang, Jiaoyang, Jagannath, Aukosh
Format:	Preprint
Publié:	2023
Sujets:	Machine Learning Probability
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Universality of high-dimensional scaling limits of stochastic gradient descent
par: Gheissari, Reza, et autres
Publié: (2025)

Local geometry of high-dimensional mixture models: Effective spectral theory and dynamical transitions
par: Arous, Gerard Ben, et autres
Publié: (2025)

Finding planted cliques using gradient descent
par: Gheissari, Reza, et autres
Publié: (2023)

Stochastic gradient descent in high dimensions for multi-spiked tensor PCA
par: Arous, Gérard Ben, et autres
Publié: (2024)

Permutation recovery of spikes in noisy high-dimensional tensor estimation
par: Arous, Gérard Ben, et autres
Publié: (2024)

Langevin dynamics for high-dimensional optimization: the case of multi-spiked tensor PCA
par: Arous, Gérard Ben, et autres
Publié: (2024)

A stochastic gradient descent algorithm with random search directions
par: Gbaguidi, Eméric
Publié: (2025)

Singular-limit analysis of gradient descent with noise injection
par: Shalova, Anna, et autres
Publié: (2024)

Pseudo-Maximum Likelihood Theory for High-Dimensional Rank One Inference
par: Grant, Curtis, et autres
Publié: (2025)

High-dimensional limit theorems for SGD: Momentum and Adaptive Step-sizes
par: Jagannath, Aukosh, et autres
Publié: (2025)

Fast Spawn\&Prune (FS\&P): Global convergence of stochastic conic particle gradient descent via birth/death process
par: De Castro, Yohann, et autres
Publié: (2026)

Optimality of Message-Passing Architectures for Sparse Graphs
par: Baranwal, Aseem, et autres
Publié: (2023)

The Larkin Mass and Replica Symmetry Breaking in the Elastic Manifold
par: Arous, Gerard Ben, et autres
Publié: (2024)

The Free Energy of the Elastic Manifold
par: Arous, Gerard Ben, et autres
Publié: (2024)

Wandering Exponents and the Free Energy of the High-Dimensional Elastic Polymer
par: Arous, Gerard Ben, et autres
Publié: (2026)

Convergence rates for gradient descent in the training of overparameterized artificial neural networks with piecewise affine activation
par: Jentzen, Arnulf, et autres
Publié: (2021)

Convergence of stochastic gradient descent schemes for Lojasiewicz-landscapes
par: Dereich, Steffen, et autres
Publié: (2021)

The generator gradient estimator is an adjoint state method for stochastic differential equations
par: Badolle, Quentin, et autres
Publié: (2024)

Mutual information and task-relevant latent dimensionality
par: Gulati, Paarth, et autres
Publié: (2026)

Exponential convergence rates for momentum stochastic gradient descent in the overparametrized setting
par: Gess, Benjamin, et autres
Publié: (2023)

Provable Benefits of Unsupervised Pre-training and Transfer Learning via Single-Index Models
par: Jones-McCormick, Taj, et autres
Publié: (2025)

Stochastic neighborhood embedding and the gradient flow of relative entropy
par: Weinkove, Ben
Publié: (2024)

Mean-field Potts and random-cluster dynamics from high-entropy initializations
par: Blanca, Antonio, et autres
Publié: (2024)

Rapid phase ordering of Ising dynamics on $\mathbb Z^2$
par: Gheissari, Reza, et autres
Publié: (2026)

Metastability in Glauber dynamics for heavy-tailed spin glasses
par: Gheissari, Reza, et autres
Publié: (2024)

Low-temperature Ising dynamics with random initializations
par: Gheissari, Reza, et autres
Publié: (2021)

Non-asymptotic convergence analysis of the stochastic gradient Hamiltonian Monte Carlo algorithm with discontinuous stochastic gradient with applications to training of ReLU neural networks
par: Liang, Luxu, et autres
Publié: (2024)

Linear convergence of proximal descent schemes on the Wasserstein space
par: Lascu, Razvan-Andrei, et autres
Publié: (2024)

A hierarchical entropy method for the delocalization of bias in high-dimensional Langevin Monte Carlo
par: Lacker, Daniel, et autres
Publié: (2025)

Langevin dynamics based algorithm e-TH$\varepsilon$O POULA for stochastic optimization problems with discontinuous stochastic gradient
par: Lim, Dong-Young, et autres
Publié: (2022)

A distance function for stochastic matrices
par: Lee, Antony R., et autres
Publié: (2024)

Mixing times of Langevin dynamics for spiked matrix models
par: Gheissari, Reza, et autres
Publié: (2026)

Convergence of gradient descent for deep neural networks
par: Chatterjee, Sourav
Publié: (2022)

On the tractability of sampling from the Potts model at low temperatures via random-cluster dynamics
par: Blanca, Antonio, et autres
Publié: (2023)

Neural stochastic Volterra equations: learning path-dependent dynamics
par: Bergerhausen, Martin, et autres
Publié: (2024)

Fisher information dissipation for time inhomogeneous stochastic differential equations
par: Feng, Qi, et autres
Publié: (2024)

Fixed-magnetization Ising on random graphs up to reconstruction
par: Gheissari, Reza, et autres
Publié: (2025)

Learning quadratic neural networks in high dimensions: SGD dynamics and scaling laws
par: Arous, Gérard Ben, et autres
Publié: (2025)

Asymptotic Gaussian Fluctuations of Eigenvectors in Spectral Clustering
par: Lebeau, Hugo, et autres
Publié: (2024)

Effective continuous equations for adaptive SGD: a stochastic analysis view
par: Callisti, Luca, et autres
Publié: (2025)