English
Related papers

Related papers: Towards Initialization-dependent and Non-vacuous G…

200 papers

Oversmoothing has been claimed as a primary bottleneck for multi-layered graph neural networks (GNNs). Multiple analyses have examined how and why oversmoothing occurs. However, none of the prior work addressed how optimization is performed…

Machine Learning · Computer Science 2024-10-08 MoonJeong Park , Dongwoo Kim

Current deep neural networks are highly overparameterized (up to billions of connection weights) and nonlinear. Yet they can fit data almost perfectly through variants of gradient descent algorithms and achieve unexpected levels of…

Explaining how overparametrized neural networks simultaneously achieve low risk and zero empirical risk on benchmark datasets is an open problem. PAC-Bayes bounds optimized using variational inference (VI) have been recently proposed as a…

Machine Learning · Computer Science 2020-03-06 Konstantinos Pitas

Recently, neural networks have demonstrated remarkable capabilities in mapping two arbitrary sets to two linearly separable sets. The prospect of achieving this with randomly initialized neural networks is particularly appealing due to the…

Machine Learning · Computer Science 2023-10-10 Promit Ghosal , Srinath Mahankali , Yihang Sun

In practice it is often found that large over-parameterized neural networks generalize better than their smaller counterparts, an observation that appears to conflict with classical notions of function complexity, which typically favor…

Machine Learning · Statistics 2018-06-20 Roman Novak , Yasaman Bahri , Daniel A. Abolafia , Jeffrey Pennington , Jascha Sohl-Dickstein

This dissertation studies a fundamental open challenge in deep learning theory: why do deep networks generalize well even while being overparameterized, unregularized and fitting the training data to zero error? In the first part of the…

Machine Learning · Computer Science 2021-10-19 Vaishnavh Nagarajan

We study over-parameterized classifiers where Empirical Risk Minimization (ERM) for learning leads to zero training error. In these over-parameterized settings there are many global minima with zero training error, some of which generalize…

Machine Learning · Computer Science 2023-12-05 Julius Martinetz , Thomas Martinetz

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

We propose a custom learning algorithm for shallow over-parameterized neural networks, i.e., networks with single hidden layer having infinite width. The infinite width of the hidden layer serves as an abstraction for the…

Machine Learning · Computer Science 2023-12-19 Alexis Teter , Iman Nodozi , Abhishek Halder

We analyze recurrent neural networks with diagonal hidden-to-hidden weight matrices, trained with gradient descent in the supervised learning setting, and prove that gradient descent can achieve optimality \emph{without} massive…

Machine Learning · Computer Science 2024-10-11 Semih Cayci , Atilla Eryilmaz

We consider the approximation rates of shallow neural networks with respect to the variation norm. Upper bounds on these rates have been established for sigmoidal and ReLU activation functions, but it has remained an important open problem…

Machine Learning · Statistics 2021-09-10 Jonathan W. Siegel , Jinchao Xu

Deep neural networks are notorious for defying theoretical treatment. However, when the number of parameters in each layer tends to infinity, the network function is a Gaussian process (GP) and quantitatively predictive description is…

Machine Learning · Computer Science 2023-10-09 Darshil Doshi , Tianyu He , Andrey Gromov

In a series of recent theoretical works, it was shown that strongly over-parameterized neural networks trained with gradient-based methods could converge exponentially fast to zero training loss, with their parameters hardly varying. In…

Optimization and Control · Mathematics 2020-01-08 Lenaic Chizat , Edouard Oyallon , Francis Bach

A recent line of work studies overparametrized neural networks in the "kernel regime," i.e. when the network behaves during training as a kernelized linear predictor, and thus training with gradient descent has the effect of finding the…

The ability of overparameterized deep networks to generalize well has been linked to the fact that stochastic gradient descent (SGD) finds solutions that lie in flat, wide minima in the training loss -- minima where the output of the…

Machine Learning · Computer Science 2019-06-03 Vaishnavh Nagarajan , J. Zico Kolter

In this paper we study the problem of learning a shallow artificial neural network that best fits a training data set. We study this problem in the over-parameterized regime where the number of observations are fewer than the number of…

Machine Learning · Computer Science 2022-08-25 Mahdi Soltanolkotabi , Adel Javanmard , Jason D. Lee

Generalization bounds which assess the difference between the true risk and the empirical risk, have been studied extensively. However, to obtain bounds, current techniques use strict assumptions such as a uniformly bounded or a Lipschitz…

Machine Learning · Computer Science 2022-11-03 Itai Gat , Yossi Adi , Alexander Schwing , Tamir Hazan

Recurrent Neural Networks (RNNs) have achieved great success in the prediction of sequential data. However, their theoretical studies are still lagging behind because of their complex interconnected structures. In this paper, we establish a…

Machine Learning · Statistics 2024-11-06 Xuewei Cheng , Ke Huang , Shujie Ma

The skip-connections used in residual networks have become a standard architecture choice in deep learning due to the increased training stability and generalization performance with this architecture, although there has been limited…

Machine Learning · Computer Science 2019-10-08 Spencer Frei , Yuan Cao , Quanquan Gu

Deep neural network (NN) with millions or billions of parameters can perform really well on unseen data, after being trained from a finite training set. Various prior theories have been developed to explain such excellent ability of NNs,…

Machine Learning · Computer Science 2025-03-11 Khoat Than , Dat Phan
‹ Prev 1 3 4 5 6 7 10 Next ›