English
Related papers

Related papers: Provable Benefit of Orthogonal Initialization in O…

200 papers

In this work, we propose a data-driven scheme to initialize the parameters of a deep neural network. This is in contrast to traditional approaches which randomly initialize parameters by sampling from transformed standard distributions.…

Neural and Evolutionary Computing · Computer Science 2021-05-24 Debasmit Das , Yash Bhalgat , Fatih Porikli

In this work, we generalize the ideas of Kaiming initialization to Graph Neural Networks (GNNs) and propose a new scheme (G-Init) that reduces oversmoothing, leading to very good results in node and graph classification tasks. GNNs are…

Machine Learning · Computer Science 2024-11-01 Dimitrios Kelesis , Dimitris Fotakis , Georgios Paliouras

In this paper, we present a novel approach to perform deep neural networks layer-wise weight initialization using Linear Discriminant Analysis (LDA). Typically, the weights of a deep neural network are initialized with: random values,…

Computer Vision and Pattern Recognition · Computer Science 2017-11-27 Michele Alberti , Mathias Seuret , Vinaychandran Pondenkandath , Rolf Ingold , Marcus Liwicki

Convergence rate of training algorithms for neural networks is heavily affected by initialization of weights. In this paper, an original algorithm for initialization of weights in backpropagation neural net is presented with application to…

Machine Learning · Computer Science 2020-04-07 Nadir Murru , Rosaria Rossini

The activation function deployed in a deep neural network has great influence on the performance of the network at initialisation, which in turn has implications for training. In this paper we study how to avoid two problems at…

Machine Learning · Computer Science 2021-05-18 Michael Murray , Vinayak Abrol , Jared Tanner

Several recent works [40, 24] observed an interesting phenomenon in neural network pruning: A larger finetuning learning rate can improve the final performance significantly. Unfortunately, the reason behind it remains elusive up to date.…

Machine Learning · Computer Science 2021-05-14 Huan Wang , Can Qin , Yue Bai , Yun Fu

A new initialization method for hidden parameters in a neural network is proposed. Derived from the integral representation of the neural network, a nonparametric probability distribution of hidden parameters is introduced. In this…

Machine Learning · Computer Science 2014-02-20 Sho Sonoda , Noboru Murata

We propose an algorithm capable of identifying and eliminating irrelevant layers of a neural network during the early stages of training. In contrast to weight or filter-level pruning, layer pruning reduces the harder to parallelize…

Machine Learning · Computer Science 2024-06-10 Valentin Frank Ingmar Guenter , Athanasios Sideris

A candidate explanation of the good empirical performance of deep neural networks is the implicit regularization effect of first order optimization methods. Inspired by this, we prove a convergence theorem for nonconvex composite…

Machine Learning · Computer Science 2023-02-14 Dávid Terjék , Diego González-Sánchez

The optimization algorithms are crucial in training physics-informed neural networks (PINNs), as unsuitable methods may lead to poor solutions. Compared to the common gradient descent (GD) algorithm, implicit gradient descent (IGD)…

Machine Learning · Computer Science 2025-08-04 Xianliang Xu , Ting Du , Wang Kong , Bin Shan , Ye Li , Zhongyi Huang

Contemporary wisdom based on empirical studies suggests that standard recurrent neural networks (RNNs) do not perform well on tasks requiring long-term memory. However, precise reasoning for this behavior is still unknown. This paper…

Machine Learning · Computer Science 2021-01-21 Melikasadat Emami , Mojtaba Sahraee-Ardakan , Parthe Pandit , Sundeep Rangan , Alyson K. Fletcher

While the impressive performance of modern neural networks is often attributed to their capacity to efficiently extract task-relevant features from data, the mechanisms underlying this rich feature learning regime remain elusive, with much…

Machine Learning · Computer Science 2024-10-15 Daniel Kunin , Allan Raventós , Clémentine Dominé , Feng Chen , David Klindt , Andrew Saxe , Surya Ganguli

Deep neural networks are usually initialized with random weights, with adequately selected initial variance to ensure stable signal propagation during training. However, selecting the appropriate variance becomes challenging especially as…

Machine Learning · Computer Science 2022-11-07 Jiawei Zhao , Florian Schäfer , Anima Anandkumar

Proper weight initialization prior to training has historically been one of the key factors that helped kick off the deep learning revolution. Initialization is even more crucial in "reservoir computing", where the weights of a readout…

Machine Learning · Computer Science 2026-05-12 Tommaso Fioratti , Riccardo Marcaccioli , Francesco Casola

Stochastic gradient descent with a large initial learning rate is widely used for training modern neural net architectures. Although a small initial learning rate allows for faster training and better test performance initially, the large…

Machine Learning · Computer Science 2020-04-28 Yuanzhi Li , Colin Wei , Tengyu Ma

Orthogonality is widely used for training deep neural networks (DNNs) due to its ability to maintain all singular values of the Jacobian close to 1 and reduce redundancy in representation. This paper proposes a computationally efficient and…

Computer Vision and Pattern Recognition · Computer Science 2020-04-03 Lei Huang , Li Liu , Fan Zhu , Diwen Wan , Zehuan Yuan , Bo Li , Ling Shao

We introduce a novel approach to perform first-order optimization with orthogonal and unitary constraints. This approach is based on a parametrization stemming from Lie group theory through the exponential map. The parametrization…

Machine Learning · Computer Science 2019-09-23 Mario Lezcano-Casado , David Martínez-Rubio

Overparameterization refers to the important phenomenon where the width of a neural network is chosen such that learning algorithms can provably attain zero loss in nonconvex training. The existing theory establishes such global convergence…

Machine Learning · Computer Science 2021-11-04 Chaehwan Song , Ali Ramezani-Kebrya , Thomas Pethick , Armin Eftekhari , Volkan Cevher

Motivated by the gap between theoretical optimal approximation rates of deep neural networks (DNNs) and the accuracy realized in practice, we seek to improve the training of DNNs. The adoption of an adaptive basis viewpoint of DNNs leads to…

Machine Learning · Computer Science 2019-12-11 Eric C. Cyr , Mamikon A. Gulian , Ravi G. Patel , Mauro Perego , Nathaniel A. Trask

Pretraining and fine-tuning are central stages in modern machine learning systems. In practice, feature learning plays an important role across both stages: deep neural networks learn a broad range of useful features during pretraining and…