English
Related papers

Related papers: Eigenvalues of the Hessian in Deep Learning: Singu…

200 papers

We consider $N\times N$ Hermitian or symmetric random matrices with independent entries. The distribution of the $(i,j)$-th matrix element is given by a probability measure $\nu_{ij}$ whose first two moments coincide with those of the…

Mathematical Physics · Physics 2011-11-16 Antti Knowles , Jun Yin

We analyse the eigenvectors of the adjacency matrix of a random inhomogeneous graph constructed from a specified degree sequence. We assume that the empirical degree sequence has bounded mean and variance. We show that near the edges of the…

Probability · Mathematics 2026-04-14 Thomas Buc-d'Alché , Antti Knowles

Differential learning rate (DLR), a technique that applies different learning rates to different model parameters, has been widely used in deep learning and achieved empirical success via its various forms. For example, parameter-efficient…

Machine Learning · Computer Science 2025-05-20 Shiyun Xu , Zhiqi Bu , Yiliang Zhang , Ian Barnett

This paper presents a novel holistic deep learning framework that simultaneously addresses the challenges of vulnerability to input perturbations, overparametrization, and performance instability from different train-validation splits. The…

Deep learning experiments by Cohen et al. [2021] using deterministic Gradient Descent (GD) revealed an Edge of Stability (EoS) phase when learning rate (LR) and sharpness (i.e., the largest eigenvalue of Hessian) no longer behave as in…

Machine Learning · Computer Science 2022-10-31 Sanjeev Arora , Zhiyuan Li , Abhishek Panigrahi

Deep learning has non-convex loss landscape and its optimization dynamics is hard to analyze or control. Nevertheless, the dynamics can be empirically convex-like across various tasks, models, optimizers, hyperparameters, etc. In this work,…

Machine Learning · Computer Science 2026-02-10 Zhiqi Bu , Shiyun Xu , Jialin Mao

Large-scale eigenvalue problems arise in various fields of science and engineering and demand computationally efficient solutions. In this study, we investigate the subspace approximation for parametric linear eigenvalue problems, aiming to…

This paper analyzes the convergence and generalization of training a one-hidden-layer neural network when the input features follow the Gaussian mixture model consisting of a finite number of Gaussian distributions. Assuming the labels are…

Machine Learning · Computer Science 2023-01-30 Hongkang Li , Shuai Zhang , Meng Wang

Machine learning algorithms frequently require careful tuning of model hyperparameters, regularization terms, and optimization parameters. Unfortunately, this tuning is often a "black art" that requires expert experience, unwritten rules of…

Machine Learning · Statistics 2012-08-30 Jasper Snoek , Hugo Larochelle , Ryan P. Adams

Edge networks call for communication efficient (low overhead) and robust distributed optimization (DO) algorithms. These are, in fact, desirable qualities for DO frameworks, such as federated edge learning techniques, in the presence of…

Systems and Control · Electrical Eng. & Systems 2023-05-19 Nicolò Dal Fabbro , Michele Rossi , Luca Schenato , Subhrakanti Dey

Recent advances in deep learning have pushed the performances of visual saliency models way further than it has ever been. Numerous models in the literature present new ways to design neural networks, to arrange gaze pattern data, or to…

Computer Vision and Pattern Recognition · Computer Science 2019-07-05 Alexandre Bruckert , Hamed R. Tavakoli , Zhi Liu , Marc Christie , Olivier Le Meur

In recent years, there has been growing interest in utilizing modern machine learning techniques to learn heuristic functions for forward search algorithms. Despite this, there has been little theoretical understanding of what they should…

Artificial Intelligence · Computer Science 2025-01-07 Carlos Núñez-Molina , Masataro Asai , Pablo Mesejo , Juan Fernández-Olivares

Recent experiments have shown that, often, when training a neural network with gradient descent (GD) with a step size $\eta$, the operator norm of the Hessian of the loss grows until it approximately reaches $2/\eta$, after which it…

Machine Learning · Computer Science 2024-06-07 Philip M. Long , Peter L. Bartlett

Recent works have studied implicit biases in deep learning, especially the behavior of last-layer features and classifier weights. However, they usually need to simplify the intermediate dynamics under gradient flow or gradient descent due…

Machine Learning · Computer Science 2023-12-14 Xiong Zhou , Xianming Liu , Hanzhang Wang , Deming Zhai , Junjun Jiang , Xiangyang Ji

Inspired by the importance of inhibitory and excitatory couplings in the brain, we analyze the largest eigenvalue statistics of random networks incorporating such features. We find that the largest real part of eigenvalues of a network,…

Disordered Systems and Neural Networks · Physics 2013-04-30 Sanjiv Kumar Dwivedi , Sarika Jalan

We study the distribution of hard-, soft-, and adaptive soft-thresholding estimators within a linear regression model where the number of parameters k can depend on sample size n and may diverge with n. In addition to the case of known…

Statistics Theory · Mathematics 2012-01-04 Benedikt M. Pötscher , Ulrike Schneider

We present an extensive analysis of relative deviation bounds, including detailed proofs of two-sided inequalities and their implications. We also give detailed proofs of two-sided generalization bounds that hold in the general case of…

Machine Learning · Computer Science 2016-04-06 Corinna Cortes , Spencer Greenberg , Mehryar Mohri

Recent theoretical studies have shown that heavy-tails can emerge in stochastic optimization due to `multiplicative noise', even under surprisingly simple settings, such as linear regression with Gaussian data. While these studies have…

Machine Learning · Statistics 2025-05-06 Mert Gurbuzbalaban , Yuanhan Hu , Umut Simsekli , Kun Yuan , Lingjiong Zhu

We propose a technique for calculating and understanding the eigenvalue distribution of sums of random matrices from the known distribution of the summands. The exact problem is formidably hard. One extreme approximation to the true density…

Quantum Physics · Physics 2017-10-27 Ramis Movassagh , Alan Edelman

Untrained large neural networks, just after random initialization, tend to favour a small subset of classes, assigning high predicted probabilities to these few classes and approximately zero probability to all others. This bias, termed…

Machine Learning · Computer Science 2025-11-27 Nicholas Pellegrino , David Szczecina , Paul W. Fieguth
‹ Prev 1 8 9 10 Next ›