English
Related papers

Related papers: Why Deep Jacobian Spectra Separate: Depth-Induced …

200 papers

We examine the geometry of neural network training using the Jacobian of trained network parameters with respect to their initial values. Our analysis reveals low-dimensional structure in the training process which is dependent on the input…

Machine Learning · Computer Science 2024-12-12 Nora Belrose , Adam Scherlis

Training neural networks via backpropagation is often hindered by vanishing or exploding gradients. In this work, we design architectures that mitigate these issues by analyzing and controlling the network Jacobian. We first provide a…

Machine Learning · Computer Science 2026-02-12 Alex Massucco , Davide Murari , Carola-Bibiane Schönlieb

Deep neural networks are known to suffer from exploding or vanishing gradients as depth increases, a phenomenon closely tied to the spectral behavior of the input-output Jacobian. Prior work has identified critical initialization schemes…

Machine Learning · Computer Science 2025-11-25 Benjamin Dadoun , Soufiane Hayou , Hanan Salam , Mohamed El Amine Seddik , Pierre Youssef

Recent work has shown that tight concentration of the entire spectrum of singular values of a deep network's input-output Jacobian around one at initialization can speed up learning by orders of magnitude. Therefore, to guide important…

Machine Learning · Statistics 2018-02-28 Jeffrey Pennington , Samuel S. Schoenholz , Surya Ganguli

Learning expressive probabilistic models correctly describing the data is a ubiquitous problem in machine learning. A popular approach for solving it is mapping the observations into a representation space with a simple joint distribution,…

Machine Learning · Statistics 2020-10-28 Luigi Gresele , Giancarlo Fissore , Adrián Javaloy , Bernhard Schölkopf , Aapo Hyvärinen

While backpropagation--reverse-mode automatic differentiation--has been extraordinarily successful in deep learning, it requires two passes (forward and backward) through the neural network and the storage of intermediate activations.…

Machine Learning · Computer Science 2025-11-06 Daniel Wang , Evan Markou , Dylan Campbell

The paradigm of differentiable programming has significantly enhanced the scope of machine learning via the judicious use of gradient-based optimization. However, standard differentiable programming methods (such as autodiff) typically…

It is well known that the initialization of weights in deep neural networks can have a dramatic impact on learning speed. For example, ensuring the mean squared singular value of a network's input-output Jacobian is $O(1)$ is essential for…

Machine Learning · Computer Science 2017-11-15 Jeffrey Pennington , Samuel S. Schoenholz , Surya Ganguli

Finding the optimal hyperparameters of a model can be cast as a bilevel optimization problem, typically solved using zero-order techniques. In this work we study first-order methods when the inner optimization problem is convex but…

This paper develops the angular and static-channel component of Geometric and Spectral Alignment for residual Jacobian chains. Starting from Cartan-coordinate rigidity and fitted effective-rank windows, we study how dominant singular…

Machine Learning · Computer Science 2026-05-05 Ziran Liu , Wei Wang , Jinhao Wang , Pengcheng Wang , Xinyi Sui , Cihan Ruan , Nam Ling , Wei Jiang

A promising trend in deep learning replaces traditional feedforward networks with implicit networks. Unlike traditional networks, implicit networks solve a fixed point equation to compute inferences. Solving for the fixed point varies in…

Machine Learning · Computer Science 2021-12-28 Samy Wu Fung , Howard Heaton , Qiuwei Li , Daniel McKenzie , Stanley Osher , Wotao Yin

Deep learning, with its exceptional learning capabilities and flexibility, has been widely applied in various applications. However, its black-box nature poses a significant challenge in real-time robotic applications, particularly in robot…

Robotics · Computer Science 2025-09-08 Koji Matsuno , Chien Chern Cheah

Deep equilibrium models (DEQs) have proven to be very powerful for learning data representations. The idea is to replace traditional (explicit) feedforward neural networks with an implicit fixed-point equation, which allows to decouple the…

Machine Learning · Computer Science 2023-04-25 Bac Nguyen , Lukas Mauch

The recent theoretical analysis of deep neural networks in their infinite-width limits has deepened our understanding of initialisation, feature learning, and training of those networks, and brought new practical techniques for finding…

Machine Learning · Computer Science 2024-08-23 Taeyoung Kim , Hongseok Yang

This article provides a comprehensive understanding of optimization in deep learning, with a primary focus on the challenges of gradient vanishing and gradient exploding, which normally lead to diminished model representational ability and…

Machine Learning · Computer Science 2023-11-14 Xianbiao Qi , Jianan Wang , Lei Zhang

Recent efforts in applying implicit networks to solve inverse problems in imaging have achieved competitive or even superior results when compared to feedforward networks. These implicit networks only require constant memory during…

Machine Learning · Computer Science 2024-02-06 Linghai Liu , Shuaicheng Tong , Lisa Zhao

Training large deep neural networks is resource intensive. This study investigates whether Lyapunov exponents can accelerate this process by aiding in the selection of hyperparameters. To study this I formulate an optimization problem using…

Machine Learning · Computer Science 2024-10-10 Tirthankar Mittra

Differentiable planning promises end-to-end differentiability and adaptivity. However, an issue prevents it from scaling up to larger-scale problems: they need to differentiate through forward iteration layers to compute gradients, which…

Machine Learning · Computer Science 2023-05-02 Linfeng Zhao , Huazhe Xu , Lawson L. S. Wong

During neural network training, the sharpness of the Hessian matrix of the training loss rises until training is on the edge of stability. As a result, even nonstochastic gradient descent does not accurately model the underlying dynamical…

Machine Learning · Statistics 2024-06-04 Mark Lowell , Catharine Kastner

We present two analytical formulae for estimating the sensitivity -- namely, the gradient or Jacobian -- at given realizations of an arbitrary-dimensional random vector with respect to its distributional parameters. The first formula…

Machine Learning · Statistics 2025-08-14 Pi-Yueh Chuang , Ahmed Attia , Emil Constantinescu
‹ Prev 1 2 3 10 Next ›