English
Related papers

Related papers: Deep Neural Network Initialization with Sparsity I…

200 papers

One of the mysteries in the success of neural networks is randomly initialized first order methods like gradient descent can achieve zero training loss even though the objective function is non-convex and non-smooth. This paper demystifies…

Machine Learning · Computer Science 2019-02-06 Simon S. Du , Xiyu Zhai , Barnabas Poczos , Aarti Singh

Neuromorphic hardware tends to pose limits on the connectivity of deep networks that one can run on them. But also generic hardware and software implementations of deep learning run more efficiently for sparse networks. Several methods…

Neural and Evolutionary Computing · Computer Science 2018-08-09 Guillaume Bellec , David Kappel , Wolfgang Maass , Robert Legenstein

Deep neural networks with lots of parameters are typically used for large-scale computer vision tasks such as image classification. This is a result of using dense matrix multiplications and convolutions. However, sparse computations are…

Computer Vision and Pattern Recognition · Computer Science 2016-11-22 Suraj Srinivas , Akshayvarun Subramanya , R. Venkatesh Babu

Trainings of Large Language Models are generally bottlenecked by matrix multiplications. In the Transformer architecture, a large portion of these operations happens in the Feed Forward Network (FFN), and this portion increases for larger…

Machine Learning · Computer Science 2026-02-09 Meghana Madhyastha , Daniel Haziza , Jesse Cai , Newsha Ardalani , Zhiqi Bu , Carole-Jean Wu

Given any deep fully connected neural network, initialized with random Gaussian parameters, we bound from above the quadratic Wasserstein distance between its output distribution and a suitable Gaussian process. Our explicit inequalities…

Machine Learning · Computer Science 2023-09-25 Andrea Basteri , Dario Trevisan

We provide novel guaranteed approaches for training feedforward neural networks with sparse connectivity. We leverage on the techniques developed previously for learning linear networks and show that they can also be effectively adopted to…

Machine Learning · Computer Science 2015-04-29 Hanie Sedghi , Anima Anandkumar

Recent results in nonparametric regression show that deep learning, i.e., neural network estimates with many hidden layers, are able to circumvent the so-called curse of dimensionality in case that suitable restrictions on the structure of…

Machine Learning · Statistics 2020-09-30 Michael Kohler , Sophie Langer

We consider neural networks with rational activation functions. The choice of the nonlinear activation function in deep learning architectures is crucial and heavily impacts the performance of a neural network. We establish optimal bounds…

Neural and Evolutionary Computing · Computer Science 2020-10-01 Nicolas Boullé , Yuji Nakatsukasa , Alex Townsend

Sparsity in the structure of Neural Networks can lead to less energy consumption, less memory usage, faster computation times on convenient hardware, and automated machine learning. If sparsity gives rise to certain kinds of structure, it…

Machine Learning · Computer Science 2021-07-28 Julian Stier , Harshil Darji , Michael Granitzer

The implicit bias induced by the training of neural networks has become a topic of rigorous study. In the limit of gradient flow and gradient descent with appropriate step size, it has been shown that when one trains a deep linear network…

Machine Learning · Computer Science 2022-04-27 Thien Le , Stefanie Jegelka

In this paper, we study total variation (TV)-regularized training of infinite-width shallow ReLU neural networks, formulated as a convex optimization problem over measures on the unit sphere. Our approach leverages the duality theory of…

Optimization and Control · Mathematics 2026-03-19 Leonardo Del Grande , Christoph Brune , Marcello Carioni

Activation functions are fundamental for enabling nonlinear representations in deep neural networks. However, the standard rectified linear unit (ReLU) often suffers from inactive or "dead" neurons caused by its hard zero cutoff. To address…

Machine Learning · Computer Science 2025-11-12 Md Motaleb Hossen Manik , Md Zabirul Islam , Ge Wang

In neural networks, non-linearity is introduced by activation functions. One commonly used activation function is Rectified Linear Unit (ReLU). ReLU has been a popular choice as an activation but has flaws. State-of-the-art functions like…

Machine Learning · Computer Science 2021-12-23 Advait Vagerwal

Small neural networks with a constrained number of trainable parameters, can be suitable resource-efficient candidates for many simple tasks, where now excessively large models are used. However, such models face several problems during the…

Machine Learning · Computer Science 2021-09-21 Alexander Kovalenko , Pavel Kordík , Magda Friedjungová

Recurrent Neural Networks (RNN) are widely used to solve a variety of problems and as the quantity of data and the amount of available compute have increased, so have model sizes. The number of parameters in recent state-of-the-art networks…

Machine Learning · Computer Science 2017-11-08 Sharan Narang , Erich Elsen , Gregory Diamos , Shubho Sengupta

Activation functions govern the expressivity and stability of neural networks, yet existing comparisons remain largely heuristic. We propose a rigorous framework for their classification via a nine-dimensional integral signature…

Machine Learning · Computer Science 2025-10-10 Ankur Mali , Lawrence Hall , Jake Williams , Gordon Richards

The performance of trained neural networks is robust to harsh levels of pruning. Coupled with the ever-growing size of deep learning models, this observation has motivated extensive research on learning sparse models. In this work, we focus…

Machine Learning · Computer Science 2022-11-29 Jose Gallego-Posada , Juan Ramirez , Akram Erraqabi , Yoshua Bengio , Simon Lacoste-Julien

We show implicit filter level sparsity manifests in convolutional neural networks (CNNs) which employ Batch Normalization and ReLU activation, and are trained with adaptive gradient descent techniques and L2 regularization or weight decay.…

Machine Learning · Computer Science 2019-05-14 Dushyant Mehta , Kwang In Kim , Christian Theobalt

The iterations of many first-order algorithms, when applied to minimizing common regularized regression functions, often resemble neural network layers with pre-specified weights. This observation has prompted the development of…

Machine Learning · Computer Science 2017-08-03 Hao He , Bo Xin , David Wipf

Despite impressive performance, deep neural networks require significant memory and computation costs, prohibiting their application in resource-constrained scenarios. Sparse training is one of the most common techniques to reduce these…

Machine Learning · Computer Science 2023-12-06 Bowen Lei , Dongkuan Xu , Ruqi Zhang , Shuren He , Bani K. Mallick