English
Related papers

Related papers: Backpropagation in matrix notation

200 papers

In this paper we focus on the linear algebra theory behind feedforward (FNN) and recurrent (RNN) neural networks. We review backward propagation, including backward propagation through time (BPTT). Also, we obtain a new exact expression for…

Machine Learning · Computer Science 2017-09-20 Maxim Naumov

This article studies (multilayer perceptron) neural networks with an emphasis on the transformations involved --- both forward and backward --- in order to develop a semantical/logical perspective that is in line with standard program…

Neural and Evolutionary Computing · Computer Science 2018-03-28 Bart Jacobs , David Sprunger

This paper introduces an extension of the backpropagation algorithm that enables us to have layers with constrained weights in a deep network. In particular, we make use of the Riemannian geometry and optimization techniques on matrix…

Computer Vision and Pattern Recognition · Computer Science 2016-11-21 Mehrtash Harandi , Basura Fernando

A matrix network is a family of matrices, with relatedness modeled by a weighted graph. We consider the task of completing a partially observed matrix network. We assume a novel sampling scheme where a fraction of matrices might be…

Machine Learning · Computer Science 2018-06-11 Qingyun Sun , Mengyuan Yan David Donoho , Stephen Boyd

Recomputation algorithms collectively refer to a family of methods that aims to reduce the memory consumption of the backpropagation by selectively discarding the intermediate results of the forward propagation and recomputing the discarded…

Machine Learning · Computer Science 2019-05-29 Mitsuru Kusumoto , Takuya Inoue , Gentaro Watanabe , Takuya Akiba , Masanori Koyama

Gradient descent has been a central training principle for artificial neural networks from the early beginnings to today's deep learning networks. The most common implementation is the backpropagation algorithm for training feed-forward…

Machine Learning · Computer Science 2020-06-09 Stefan Jaeger

We propose proximal backpropagation (ProxProp) as a novel algorithm that takes implicit instead of explicit gradient steps to update the network parameters during neural network training. Our algorithm is motivated by the step size…

Machine Learning · Computer Science 2018-02-21 Thomas Frerix , Thomas Möllenhoff , Michael Moeller , Daniel Cremers

The goal of this document is to provide a pedagogical introduction to the main concepts underpinning the training of deep neural networks using gradient descent; a process known as backpropagation. Although we focus on a very influential…

Machine Learning · Computer Science 2018-11-30 Laurent Boué

We introduce a novel method to compute a rank $m$ approximation of the inverse of the Hessian matrix in the distributed regime. By leveraging the differences in gradients and parameters of multiple Workers, we are able to efficiently…

Machine Learning · Computer Science 2017-09-18 Sébastien M. R. Arnold , Chunming Wang

In this paper, we utilize weight gradients from backpropagation to characterize the representation space learned by deep learning algorithms. We demonstrate the utility of such gradients in applications including perceptual image quality…

Computer Vision and Pattern Recognition · Computer Science 2019-08-28 Gukyeong Kwon , Mohit Prabhushankar , Dogancan Temel , Ghassan AlRegib

The backpropagation algorithm for calculating gradients has been widely used in computation of weights for deep neural networks (DNNs). This method requires derivatives of objective functions and has some difficulties finding appropriate…

Machine Learning · Computer Science 2016-05-17 Tetsuya Sakurai , Akira Imakura , Yuto Inoue , Yasunori Futamura

Deep residual networks (ResNets) have significantly pushed forward the state-of-the-art on image classification, increasing in performance as networks grow both deeper and wider. However, memory consumption becomes a bottleneck, as one…

Computer Vision and Pattern Recognition · Computer Science 2017-07-18 Aidan N. Gomez , Mengye Ren , Raquel Urtasun , Roger B. Grosse

This paper establishes risk convergence and asymptotic weight matrix alignment --- a form of implicit regularization --- of gradient flow and gradient descent when applied to deep linear networks on linearly separable data. In more detail,…

Machine Learning · Computer Science 2019-02-26 Ziwei Ji , Matus Telgarsky

Neural networks are trained primarily based on their inputs and outputs, without regard for their internal mechanisms. These neglected mechanisms determine properties that are critical for safety, like (i) transparency; (ii) the absence of…

Machine Learning · Computer Science 2024-12-02 Alex Cloud , Jacob Goldman-Wetzler , Evžen Wybitul , Joseph Miller , Alexander Matt Turner

Memory footprint is one of the main limiting factors for large neural network training. In backpropagation, one needs to store the input to each operation in the computational graph. Every modern neural network model has quite a few…

Machine Learning · Computer Science 2022-02-04 Georgii Novikov , Daniel Bershatsky , Julia Gusak , Alex Shonenkov , Denis Dimitrov , Ivan Oseledets

This technical report describes an efficient technique for computing the norm of the gradient of the loss function for a neural network with respect to its parameters. This gradient norm can be computed efficiently for every example.

Machine Learning · Statistics 2015-10-13 Ian Goodfellow

As deep learning applications continue to deploy increasingly large artificial neural networks, the associated high energy demands are creating a need for alternative neuromorphic approaches. Optics and photonics are particularly compelling…

Optics · Physics 2025-08-19 Nicola Dal Cin , Florian Marquardt , Clara C. Wanjura

We study a class of deep neural networks with networks that form a directed acyclic graph (DAG). For backpropagation defined by gradient descent with adaptive momentum, we show weights converge for a large class of nonlinear activation…

Computer Vision and Pattern Recognition · Computer Science 2020-01-22 Chirag Agarwal , Joe Klobusicky , Dan Schonfeld

Recently, integrated optics has gained interest as a hardware platform for implementing machine learning algorithms. Of particular interest are artificial neural networks, since matrix-vector multi- plications, which are used heavily in…

Optics · Physics 2018-07-25 Tyler W. Hughes , Momchil Minkov , Yu Shi , Shanhui Fan

The shrinkage function is widely used in matrix low-rank approximation, compressive sensing, and statistical estimation. In this article, an elementary derivation of the shrinkage function is given. In addition, applications of the…

Optimization and Control · Mathematics 2017-03-30 Toby Boas , Aritra Dutta , Xin Li , Kathryn P. Mercier , Eric Niderman