中文
相关论文

相关论文: Collapse of Deep and Narrow Neural Nets

200 篇论文

Neural Collapse is a phenomenon where the last-layer representations of a well-trained neural network converge to a highly structured geometry. In this paper, we focus on its first (and most basic) property, known as NC1: the within-class…

机器学习 · 计算机科学 2025-02-05 Diyuan Wu , Marco Mondelli

This work investigates the generalization behavior of deep neural networks (DNNs), focusing on the phenomenon of "fooling examples," where DNNs confidently classify inputs that appear random or unstructured to humans. To explore this…

机器学习 · 计算机科学 2025-08-22 Yen-Lung Lai , Zhe Jin

Neural Collapse is a phenomenon that helps identify sparse and low rank structures in deep classifiers. Recent work has extended the definition of neural collapse to regression problems, albeit only measuring the phenomenon at the last…

机器学习 · 计算机科学 2026-03-26 Akshay Rangamani , Altay Unal

This work studies the behavior of shallow ReLU networks trained with the logistic loss via gradient descent on binary classification data where the underlying data distribution is general, and the (optimal) Bayes risk is not necessarily…

机器学习 · 计算机科学 2021-11-05 Ziwei Ji , Justin D. Li , Matus Telgarsky

This article contributes to the current statistical theory of deep neural networks (DNNs). It was shown that DNNs are able to circumvent the so--called curse of dimensionality in case that suitable restrictions on the structure of the…

统计理论 · 数学 2020-10-14 Sophie Langer

We study the theory of neural network (NN) from the lens of classical nonparametric regression problems with a focus on NN's ability to adaptively estimate functions with heterogeneous smoothness -- a property of functions in Besov or…

机器学习 · 计算机科学 2024-05-21 Kaiqi Zhang , Yu-Xiang Wang

Convolutional and Recurrent, deep neural networks have been successful in machine learning systems for computer vision, reinforcement learning, and other allied fields. However, the robustness of such neural networks is seldom apprised,…

神经与进化计算 · 计算机科学 2018-05-01 Biswa Sengupta , Karl J. Friston

Deep learning relies on good initialization schemes and hyperparameter choices prior to training a neural network. Random weight initializations induce random network ensembles, which give rise to the trainability, training speed, and…

机器学习 · 统计学 2019-10-25 Rebekka Burkholz , Alina Dubatovka

One fundamental problem in deep learning is understanding the outstanding performance of deep Neural Networks (NNs) in practice. One explanation for the superiority of NNs is that they can realize a large class of complicated functions,…

机器学习 · 计算机科学 2020-06-30 H. Xiong , L. Huang , M. Yu , L. Liu , F. Zhu , L. Shao

A main theoretical interest in biology and physics is to identify the nonlinear dynamical system (DS) that generated observed time series. Recurrent Neural Networks (RNNs) are, in principle, powerful enough to approximate any underlying DS,…

机器学习 · 计算机科学 2021-03-15 Dominik Schmidt , Georgia Koppe , Zahra Monfared , Max Beutelspacher , Daniel Durstewitz

Despite the efficacy on a variety of computer vision tasks, deep neural networks (DNNs) are vulnerable to adversarial attacks, limiting their applications in security-critical systems. Recent works have shown the possibility of generating…

计算机视觉与模式识别 · 计算机科学 2018-12-21 Ziang Yan , Yiwen Guo , Changshui Zhang

The deep neural networks (DNNs) have achieved great success in learning complex patterns with strong predictive power, but they are often thought of as "black box" models without a sufficient level of transparency and interpretability. It…

机器学习 · 计算机科学 2020-11-10 Agus Sudjianto , William Knauth , Rahul Singh , Zebin Yang , Aijun Zhang

Activation functions influence behavior and performance of DNNs. Nonlinear activation functions, like Rectified Linear Units (ReLU), Exponential Linear Units (ELU) and Scaled Exponential Linear Units (SELU), outperform the linear…

神经与进化计算 · 计算机科学 2019-02-05 Alberto Marchisio , Muhammad Abdullah Hanif , Semeen Rehman , Maurizio Martina , Muhammad Shafique

Although deep learning has shown its powerful performance in many applications, the mathematical principles behind neural networks are still mysterious. In this paper, we consider the problem of learning a one-hidden-layer neural network…

机器学习 · 计算机科学 2019-07-17 Shuhao Xia , Yuanming Shi

Generalization of deep neural networks remains one of the main open problems in machine learning. Previous theoretical works focused on deriving tight bounds of model complexity, while empirical works revealed that neural networks exhibit…

机器学习 · 计算机科学 2022-01-31 James Wang , Cheng-Lin Yang

Deep learning algorithms -- typically consisting of a class of deep artificial neural networks (ANNs) trained by a stochastic gradient descent (SGD) optimization method -- are nowadays an integral part in many areas of science, industry,…

机器学习 · 计算机科学 2025-01-28 Steffen Dereich , Thang Do , Arnulf Jentzen , Frederic Weber

Deep Neural Networks (DNNs) are rapidly being applied to safety-critical domains such as drone and airplane control, motivating techniques for verifying the safety of their behavior. Unfortunately, DNN verification is NP-hard, with current…

机器学习 · 计算机科学 2020-09-15 Matthew Sotoudeh , Aditya V. Thakur

Rectified Linear Units (ReLU) are the default choice for activation functions in deep neural networks. While they demonstrate excellent empirical performance, ReLU activations can fall victim to the dead neuron problem. In these cases, the…

机器学习 · 计算机科学 2023-02-14 Tim Whitaker , Darrell Whitley

The current paradigm of training deep neural networks for classification tasks includes minimizing the empirical risk that pushes the training loss value towards zero, even after the training error has been vanished. In this terminal phase…

机器学习 · 计算机科学 2024-06-07 Hien Dang , Tho Tran , Tan Nguyen , Nhat Ho

We investigate the loss surface of neural networks. We prove that even for one-hidden-layer networks with "slightest" nonlinearity, the empirical risks have spurious local minima in most cases. Our results thus indicate that in general "no…

机器学习 · 计算机科学 2019-05-29 Chulhee Yun , Suvrit Sra , Ali Jadbabaie