English
Related papers

Related papers: Functional Large Deviations for Wide Deep Neural N…

200 papers

The problem of quantizing the activations of a deep neural network is considered. An examination of the popular binary quantization approach shows that this consists of approximating a classical non-linearity, the hyperbolic tangent, by two…

Computer Vision and Pattern Recognition · Computer Science 2017-02-06 Zhaowei Cai , Xiaodong He , Jian Sun , Nuno Vasconcelos

We establish the large deviation principle for solutions of one-dimensional SDEs with discontinuous coefficients. The main statement is formulated in a form similar to the classical Wentzel--Freidlin theorem, but under the considerably…

Probability · Mathematics 2016-07-14 Alexei Kulik , Daryna Sobolieva

Covering numbers of (deep) ReLU networks have been used to characterize approximation-theoretic performance, to upper-bound prediction error in nonparametric regression, and to quantify classification capacity. These results rely on…

Machine Learning · Statistics 2026-03-04 Weigutian Ou , Helmut Bölcskei

Artificial neural networks (NNs) have become the de facto standard in machine learning. They allow learning highly nonlinear transformations in a plethora of applications. However, NNs usually only provide point estimates without…

Machine Learning · Statistics 2020-09-11 Marco F. Huber

In this paper, we study the large deviation principle of invariant measures of stochastic reaction-diffusion lattice systems driven by multiplicative noise. We first show that any limit of a sequence of invariant measures of the stochastic…

Probability · Mathematics 2024-05-07 Bixiang Wang

We investigate the convergence guarantee of two-layer neural network training with Gaussian randomly masked inputs. This scenario corresponds to Gaussian dropout at the input level, or noisy input training common in sensor networks,…

Machine Learning · Computer Science 2026-02-20 Afroditi Kolomvaki , Fangshuo Liao , Evan Dramko , Ziyun Guang , Anastasios Kyrillidis

The local Lipschitz constant of a neural network is a useful metric with applications in robustness, generalization, and fairness evaluation. We provide novel analytic results relating the local Lipschitz constant of nonsmooth vector-valued…

Machine Learning · Statistics 2021-01-12 Matt Jordan , Alexandros G. Dimakis

In a series of two papers, we investigate the large deviations and asymptotic behavior of stochastic models of brain neural networks with random interaction coefficients. In this first paper, we take into account the spatial structure of…

Probability · Mathematics 2017-01-05 Tanguy Cabana , Jonathan Touboul

Activation sparsity is an intriguing property of deep neural networks that has been extensively studied in ReLU-based models, due to its advantages for efficiency, robustness, and interpretability. However, methods relying on exact zero…

We study the computation complexity of deep ReLU (Rectified Linear Unit) neural networks for the approximation of functions from the H\"older-Zygmund space of mixed smoothness defined on the $d$-dimensional unit cube when the dimension $d$…

Numerical Analysis · Mathematics 2021-07-26 Dinh Dũng , Van Kien Nguyen

Neural networks with random weights appear in a variety of machine learning applications, most prominently as the initialization of many deep learning algorithms and as a computationally cheap alternative to fully learned neural networks.…

Machine Learning · Computer Science 2022-11-29 Sjoerd Dirksen , Martin Genzel , Laurent Jacques , Alexander Stollenwerk

Implicit deep learning has received increasing attention recently due to the fact that it generalizes the recursive prediction rules of many commonly used neural network architectures. Its prediction rule is provided implicitly based on the…

Machine Learning · Computer Science 2022-02-21 Tianxiang Gao , Hailiang Liu , Jia Liu , Hridesh Rajan , Hongyang Gao

We present empirical evidence that neural networks with ReLU and Absolute Value activations learn distance-based representations. We independently manipulate both distance and intensity properties of internal activations in trained models,…

Machine Learning · Computer Science 2024-11-28 Alan Oursland

Motivated by the growing theoretical understanding of neural networks that employ the Rectified Linear Unit (ReLU) as their activation function, we revisit the use of ReLU activation functions for learning implicit neural representations…

Image and Video Processing · Electrical Eng. & Systems 2024-08-05 Joseph Shenouda , Yamin Zhou , Robert D. Nowak

Activation in deep neural networks is fundamental to achieving non-linear mappings. Traditional studies mainly focus on finding fixed activations for a particular set of learning tasks or model architectures. The research on flexible…

Neural and Evolutionary Computing · Computer Science 2020-08-20 Renlong Jie , Junbin Gao , Andrey Vasnev , Min-ngoc Tran

Activation function is crucial to the recent successes of deep neural networks. In this paper, we first propose a new activation function, Multiple Parametric Exponential Linear Units (MPELU), aiming to generalize and unify the rectified…

Computer Vision and Pattern Recognition · Computer Science 2017-01-18 Yang Li , Chunxiao Fan , Yong Li , Qiong Wu , Yue Ming

We study the large deviation behaviour of the trajectories of empirical distributions of independent copies of time-homogeneous Feller processes on locally compact metric spaces. Under the condition that we can find a suitable core for the…

Functional Analysis · Mathematics 2018-03-13 Richard C. Kraaij

We study wide Bayesian neural networks focusing on the rare but statistically dominant fluctuations that govern posterior concentration, beyond Gaussian-process limits. Large-deviation theory provides explicit variational objectives-rate…

Machine Learning · Statistics 2026-02-27 Katerina Papagiannouli , Dario Trevisan , Giuseppe Pio Zitto

We present a novel approach to implementing all-optical Rectified Linear Unit (ReLU) activation functions using compact doubly-resonant cavities with dimensions of approximately $10\,\mu\mathrm{m}$. Our design leverages $\chi^{(2)}$…

Optics · Physics 2025-04-29 Amirreza Ahmadnejad , Mohmmad Mehrdad Asadi , Somayyeh Koohi

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