English
Related papers

Related papers: The exploding gradient problem demystified - defin…

200 papers

"Deep Learning"/"Deep Neural Nets" is a technological marvel that is now increasingly deployed at the cutting-edge of artificial intelligence tasks. This dramatic success of deep learning in the last few years has been hinged on an enormous…

Machine Learning · Computer Science 2021-04-30 Anirbit Mukherjee

We prove that two-layer (Leaky)ReLU networks initialized by e.g. the widely used method proposed by He et al. (2015) and trained using gradient descent on a least-squares loss are not universally consistent. Specifically, we describe a…

Machine Learning · Statistics 2022-06-10 David Holzmüller , Ingo Steinwart

Attention and self-attention mechanisms, are now central to state-of-the-art deep learning on sequential tasks. However, most recent progress hinges on heuristic approaches with limited understanding of attention's role in model…

Machine Learning · Computer Science 2020-12-11 Giancarlo Kerg , Bhargav Kanuparthi , Anirudh Goyal , Kyle Goyette , Yoshua Bengio , Guillaume Lajoie

A recent line of research on deep learning focuses on the extremely over-parameterized setting, and shows that when the network width is larger than a high degree polynomial of the training sample size $n$ and the inverse of the target…

Machine Learning · Computer Science 2022-01-03 Zixiang Chen , Yuan Cao , Difan Zou , Quanquan Gu

The unprecedented success of deep learning (DL) makes it unchallenged when it comes to classification problems. However, it is well established that the current DL methodology produces universally unstable neural networks (NNs). The…

Machine Learning · Computer Science 2025-03-27 Alexander Bastounis , Anders C Hansen , Verner Vlačić

Stochastic gradient descent (SGD) optimization methods such as the plain vanilla SGD method and the popular Adam optimizer are nowadays the method of choice in the training of artificial neural networks (ANNs). Despite the remarkable…

Optimization and Control · Mathematics 2024-02-09 Arnulf Jentzen , Adrian Riekert

Training a single network with multiple objectives often leads to conflicting gradients that degrade shared representations, forcing them into a compromised state that is suboptimal for any single task--a problem we term latent…

Machine Learning · Computer Science 2026-03-03 Chi-Yao Huang , Khoa Vo , Aayush Atul Verma , Duo Lu , Yezhou Yang

Understanding implicit bias of gradient descent for generalization capability of ReLU networks has been an important research topic in machine learning research. Unfortunately, even for a single ReLU neuron trained with the square loss, it…

Machine Learning · Computer Science 2022-06-14 Sangmin Lee , Byeongsu Sim , Jong Chul Ye

We study how neural networks trained by gradient descent extrapolate, i.e., what they learn outside the support of the training distribution. Previous works report mixed empirical results when extrapolating with neural networks: while…

Machine Learning · Computer Science 2021-03-04 Keyulu Xu , Mozhi Zhang , Jingling Li , Simon S. Du , Ken-ichi Kawarabayashi , Stefanie Jegelka

Convergence of deep neural networks as the depth of the networks tends to infinity is fundamental in building the mathematical foundation for deep learning. In a previous study, we investigated this question for deep ReLU networks with a…

Machine Learning · Computer Science 2022-01-25 Yuesheng Xu , Haizhang Zhang

We draw connections between simple neural networks and under-determined linear systems to comprehensively explore several interesting theoretical questions in the study of neural networks. First, we emphatically show that it is unsurprising…

Numerical Analysis · Mathematics 2020-11-02 Austin R. Benson , Anil Damle , Alex Townsend

The training of artificial neural networks (ANNs) with rectified linear unit (ReLU) activation via gradient descent (GD) type optimization schemes is nowadays a common industrially relevant procedure. Till this day in the scientific…

Machine Learning · Computer Science 2023-04-13 Simon Eberle , Arnulf Jentzen , Adrian Riekert , Georg S. Weiss

We study randomly initialized residual networks using mean field theory and the theory of difference equations. Classical feedforward neural networks, such as those with tanh activations, exhibit exponential behavior on the average when…

Neural and Evolutionary Computing · Computer Science 2017-12-27 Greg Yang , Samuel S. Schoenholz

Gradient coding is a technique for straggler mitigation in distributed learning. In this paper we design novel gradient codes using tools from classical coding theory, namely, cyclic MDS codes, which compare favorably with existing…

Information Theory · Computer Science 2019-07-09 Netanel Raviv , Itzhak Tamo , Rashish Tandon , Alexandros G. Dimakis

Deep learning algorithms -- typically consisting of a class of deep artificial neural networks (ANNs) trained by a stochastic gradient descent (SGD) optimization method -- are nowadays an integral part in many areas of science, industry,…

Machine Learning · Computer Science 2025-01-28 Steffen Dereich , Thang Do , Arnulf Jentzen , Frederic Weber

We provide a theoretical explanation for the effectiveness of gradient clipping in training deep neural networks. The key ingredient is a new smoothness condition derived from practical neural network training examples. We observe that…

Optimization and Control · Mathematics 2020-02-12 Jingzhao Zhang , Tianxing He , Suvrit Sra , Ali Jadbabaie

This paper critically examines the fundamental distinctions between gradient methods applied to non-differentiable functions (NGDMs) and classical gradient descents (GDs) for differentiable functions, revealing significant gaps in current…

Machine Learning · Computer Science 2025-07-31 Siddharth Krishna Kumar

It is well known that the problem of vanishing/exploding gradients is a challenge when training deep networks. In this paper, we describe another phenomenon, called vanishing nodes, that also increases the difficulty of training deep neural…

Machine Learning · Computer Science 2019-10-23 Wen-Yu Chang , Tsung-Nan Lin

Implicit neural networks have become increasingly attractive in the machine learning community since they can achieve competitive performance but use much less computational resources. Recently, a line of theoretical works established the…

Machine Learning · Computer Science 2022-10-03 Tianxiang Gao , Hongyang Gao

Randomly initialized neural networks are known to become harder to train with increasing depth, unless architectural enhancements like residual connections and batch normalization are used. We here investigate this phenomenon by revisiting…

Machine Learning · Statistics 2020-06-15 Hadi Daneshmand , Jonas Kohler , Francis Bach , Thomas Hofmann , Aurelien Lucchi