English
Related papers

Related papers: Reintroducing Straight-Through Estimators as Princ…

200 papers

This work develops a new method for estimating and optimizing the directed information rate between two jointly stationary and ergodic stochastic processes. Building upon recent advances in machine learning, we propose a recurrent neural…

Information Theory · Computer Science 2022-03-29 Dor Tsur , Ziv Aharoni , Ziv Goldfeld , Haim Permuter

Turning the weights to zero when training a neural network helps in reducing the computational complexity at inference. To progressively increase the sparsity ratio in the network without causing sharp weight discontinuities during…

Computer Vision and Pattern Recognition · Computer Science 2023-01-25 Antoine Vanderschueren , Christophe De Vleeschouwer

In this paper we introduce a family of stochastic gradient estimation techniques based of the perturbative expansion around the mean of the sampling distribution. We characterize the bias and variance of the resulting Taylor-corrected…

Machine Learning · Statistics 2019-11-18 Luca Ambrogioni , Marcel A. J. van Gerven

The back-propagation (BP) algorithm has been considered the de-facto method for training deep neural networks. It back-propagates errors from the output layer to the hidden layers in an exact manner using the transpose of the feedforward…

Neural and Evolutionary Computing · Computer Science 2018-05-01 Hongyin Luo , Jie Fu , James Glass

We explore training Binary Neural Networks (BNNs) as a discrete variable inference problem over a factor graph. We study the behaviour of this conversion in an under-parameterized BNN setting and propose stochastic versions of Belief…

Machine Learning · Computer Science 2022-04-06 Amir Khoshaman , Giuseppe Castiglione , Christopher Srinivasa

We introduce a method to train Binarized Neural Networks (BNNs) - neural networks with binary weights and activations at run-time. At training-time the binary weights and activations are used for computing the parameters gradients. During…

Machine Learning · Computer Science 2016-03-18 Matthieu Courbariaux , Itay Hubara , Daniel Soudry , Ran El-Yaniv , Yoshua Bengio

This paper proposes a novel binarized weight network (BT) for a resource-efficient neural structure. The proposed model estimates a binary representation of weights by taking into account the approximation error with an additional term.…

Computer Vision and Pattern Recognition · Computer Science 2021-05-11 Savas Ozkan , Gozde Bozdagi Akar

Spiking Neural Networks (SNNs) are a promising approach to low-power applications on neuromorphic hardware due to their energy efficiency. However, training SNNs is challenging because of the non-differentiable spike generation function. To…

Neural and Evolutionary Computing · Computer Science 2025-08-19 Qingyan Meng , Mingqing Xiao , Zhengyu Ma , Huihui Zhou , Yonghong Tian , Zhouchen Lin

The Straight-Through (ST) estimator is a widely used technique for back-propagating gradients through discrete random variables. However, this effective method lacks theoretical justification. In this paper, we show that ST can be…

Machine Learning · Computer Science 2019-10-08 Pengyu Cheng , Chang Liu , Chunyuan Li , Dinghan Shen , Ricardo Henao , Lawrence Carin

Recurrent Neural Networks (RNNs) produce state-of-art performance on many machine learning tasks but their demand on resources in terms of memory and computational power are often high. Therefore, there is a great interest in optimizing the…

Neural and Evolutionary Computing · Computer Science 2017-02-28 Joachim Ott , Zhouhan Lin , Ying Zhang , Shih-Chii Liu , Yoshua Bengio

During the last few years, significant attention has been paid to the stochastic training of artificial neural networks, which is known as an effective regularization approach that helps improve the generalization capability of trained…

Machine Learning · Computer Science 2018-12-04 Qi Sun , Yunzhe Tao , Qiang Du

Stochastic nonlinear dynamical systems are ubiquitous in modern, real-world applications. Yet, estimating the unknown parameters of stochastic, nonlinear dynamical models remains a challenging problem. The majority of existing methods…

Machine Learning · Statistics 2022-05-06 Anubhab Ghosh , Mohamed Abdalmoaty , Saikat Chatterjee , Håkan Hjalmarsson

Recent research has shown that one can train a neural network with binary weights and activations at train time by augmenting the weights with a high-precision continuous latent variable that accumulates small changes from stochastic…

Machine Learning · Computer Science 2017-05-23 Alexander G. Anderson , Cory P. Berg

Spiking Neural Networks (SNNs) emerged as a promising solution in the field of Artificial Neural Networks (ANNs), attracting the attention of researchers due to their ability to mimic the human brain and process complex information with…

Neural and Evolutionary Computing · Computer Science 2023-09-12 Pavithra Koralalage , Ireoluwa Fakeye , Pedro Machado , Jason Smith , Isibor Kennedy Ihianle , Salisu Wada Yahaya , Andreas Oikonomou , Ahmad Lotfi

Spatial Transformer Networks (STNs) estimate image transformations that can improve downstream tasks by `zooming in' on relevant regions in an image. However, STNs are hard to train and sensitive to mis-predictions of transformations. To…

Machine Learning · Computer Science 2022-06-16 Pola Schwöbel , Frederik Warburg , Martin Jørgensen , Kristoffer H. Madsen , Søren Hauberg

Spiking neural networks (SNNs) represent a promising approach in machine learning, combining the hierarchical learning capabilities of deep neural networks with the energy efficiency of spike-based computations. Traditional end-to-end…

Neural and Evolutionary Computing · Computer Science 2024-11-12 Ruyin Wan , Qian Zhang , George Em Karniadakis

Binary Neural Networks (BNNs) are an extremely promising method to reduce deep neural networks' complexity and power consumption massively. Binarization techniques, however, suffer from ineligible performance degradation compared to their…

Machine Learning · Computer Science 2022-04-06 Tal Rozen , Moshe Kimhi , Brian Chmiel , Avi Mendelson , Chaim Baskin

The Straight-Through Estimator (STE) is the dominant method for training neural networks with discrete variables, enabling gradient-based optimisation by routing gradients through a differentiable surrogate. However, existing STE variants…

Machine Learning · Computer Science 2026-02-24 Rushi Shah , Mingyuan Yan , Michael Curtis Mozer , Dianbo Liu

Stochastic binary hidden units in a multi-layer perceptron (MLP) network give at least three potential benefits when compared to deterministic MLP networks. (1) They allow to learn one-to-many type of mappings. (2) They can be used in…

Machine Learning · Statistics 2015-04-10 Tapani Raiko , Mathias Berglund , Guillaume Alain , Laurent Dinh

A plethora of recent research has focused on improving the memory footprint and inference speed of deep networks by reducing the complexity of (i) numerical representations (for example, by deterministic or stochastic quantization) and (ii)…

Machine Learning · Computer Science 2019-04-05 David Hartmann , Michael Wand