English
Related papers

Related papers: Symmetry-invariant optimization in deep networks

200 papers

Stochastic gradient descent (SGD) is the main approach for training deep networks: it moves towards the optimum of the cost function by iteratively updating the parameters of a model in the direction of the gradient of the loss evaluated on…

Machine Learning · Computer Science 2021-03-30 Loris Nanni , Gianluca Maguolo , Alessandra Lumini

We introduce data structures for solving robust regression through stochastic gradient descent (SGD) by sampling gradients with probability proportional to their norm, i.e., importance sampling. Although SGD is widely used for large scale…

Machine Learning · Computer Science 2022-07-19 Sepideh Mahabadi , David P. Woodruff , Samson Zhou

-The fluctuation effect of gradient expectation and variance caused by parameter update between consecutive iterations is neglected or confusing by current mainstream gradient optimization algorithms. The work in this paper remedy this…

Machine Learning · Statistics 2021-10-08 Aixiang Chen

We study deep neural networks for classification of images with quality distortions. We first show that networks fine-tuned on distorted data greatly outperform the original networks when tested on distorted data. However, fine-tuned…

Computer Vision and Pattern Recognition · Computer Science 2017-03-24 Samuel Dodge , Lina Karam

Using weight decay to penalize the L2 norms of weights in neural networks has been a standard training practice to regularize the complexity of networks. In this paper, we show that a family of regularizers, including weight decay, is…

Machine Learning · Computer Science 2022-06-09 Ziquan Liu , Yufei Cui , Antoni B. Chan

Spectral gradient methods, such as the Muon optimizer, modify gradient updates by preserving directional information while discarding scale, and have shown strong empirical performance in deep learning. We investigate the mechanisms…

Machine Learning · Statistics 2026-02-02 Guillaume Braun , Han Bao , Wei Huang , Masaaki Imaizumi

The energy landscape of high-dimensional non-convex optimization problems is crucial to understanding the effectiveness of modern deep neural network architectures. Recent works have experimentally shown that two different solutions found…

Machine Learning · Computer Science 2024-03-04 Damien Ferbach , Baptiste Goujaud , Gauthier Gidel , Aymeric Dieuleveut

We study the convergence of gradient descent (GD) and stochastic gradient descent (SGD) for training $L$-hidden-layer linear residual networks (ResNets). We prove that for training deep residual networks with certain linear transformations…

Machine Learning · Computer Science 2020-03-03 Difan Zou , Philip M. Long , Quanquan Gu

Bilevel optimization is a central tool in machine learning for high-dimensional hyperparameter tuning. Its applications are vast; for instance, in imaging it can be used for learning data-adaptive regularizers and optimizing forward…

Optimization and Control · Mathematics 2025-11-11 Mohammad Sadegh Salehi , Subhadip Mukherjee , Lindon Roberts , Matthias J. Ehrhardt

Implicit bias induced by gradient-based algorithms is essential to the generalization of overparameterized models, yet its mechanisms can be subtle. This work leverages the Normalized Steepest Descent} (NSD) framework to investigate how…

Machine Learning · Computer Science 2026-03-25 Shengping Xie , Zekun Wu , Quan Chen , Kaixu Tang

Ongoing studies have identified similarities between neural representations in biological networks and in deep artificial neural networks. This has led to renewed interest in developing analogies between the backpropagation learning…

Neural and Evolutionary Computing · Computer Science 2019-06-11 Theodore H. Moskovitz , Ashok Litwin-Kumar , L. F. Abbott

Neural networks are usually trained by some form of stochastic gradient descent (SGD)). A number of strategies are in common use intended to improve SGD optimization, such as learning rate schedules, momentum, and batching. These are…

Neural and Evolutionary Computing · Computer Science 2015-08-13 Thomas M. Breuel

Current state-of-the-art deep neural networks for image classification are made up of 10 - 100 million learnable weights and are therefore inherently prone to overfitting. The complexity of the weight count can be seen as a function of the…

Computer Vision and Pattern Recognition · Computer Science 2022-11-11 Antonia van Betteray , Matthias Rottmann , Karsten Kahl

Variational approaches to disparity estimation typically use a linearised brightness constancy constraint, which only applies in smooth regions and over small distances. Accordingly, current variational approaches rely on a schedule to…

Image and Video Processing · Electrical Eng. & Systems 2024-05-28 James L. Gray , Aous T. Naman , David S. Taubman

We report, for the first time, on the cascade weight shedding phenomenon in deep neural networks where in response to pruning a small percentage of a network's weights, a large percentage of the remaining is shed over a few epochs during…

Machine Learning · Computer Science 2021-03-22 Kambiz Azarian , Fatih Porikli

With the rise of big data analytics, multi-layer neural networks have surfaced as one of the most powerful machine learning methods. However, their theoretical mathematical properties are still not fully understood. Training a neural…

Machine Learning · Computer Science 2021-01-01 Victor Luo , Yazhen Wang , Glenn Fung

Recent advances in reconstruction methods for inverse problems leverage powerful data-driven models, e.g., deep neural networks. These techniques have demonstrated state-of-the-art performances for several imaging tasks, but they often do…

Computer Vision and Pattern Recognition · Computer Science 2020-10-20 Riccardo Barbano , Chen Zhang , Simon Arridge , Bangti Jin

Many types of neural network layers rely on matrix properties such as invertibility or orthogonality. Retaining such properties during optimization with gradient-based stochastic optimizers is a challenging task, which is usually addressed…

Machine Learning · Statistics 2020-12-02 Andreas Krämer , Jonas Köhler , Frank Noé

Deep neural networks are typically trained by optimizing a loss function with an SGD variant, in conjunction with a decaying learning rate, until convergence. We show that simple averaging of multiple points along the trajectory of SGD,…

Machine Learning · Computer Science 2019-02-26 Pavel Izmailov , Dmitrii Podoprikhin , Timur Garipov , Dmitry Vetrov , Andrew Gordon Wilson

Stochastic gradient descent (SGD) is a standard optimization method to minimize a training error with respect to network parameters in modern neural network learning. However, it typically suffers from proliferation of saddle points in the…

Machine Learning · Computer Science 2017-11-23 Haiping Huang , Taro Toyoizumi