中文
相关论文

相关论文: Evolution of Neural Tangent Kernels under Benign a…

200 篇论文

The ability of learning useful features is one of the major advantages of neural networks. Although recent works show that neural network can operate in a neural tangent kernel (NTK) regime that does not allow feature learning, many works…

机器学习 · 计算机科学 2024-11-06 Mo Zhou , Rong Ge

Decentralized federated learning (DFL) is a collaborative machine learning framework for training a model across participants without a central server or raw data exchange. DFL faces challenges due to statistical heterogeneity, as…

机器学习 · 计算机科学 2025-06-16 Gabriel Thompson , Kai Yue , Chau-Wai Wong , Huaiyu Dai

A key property of neural networks driving their success is their ability to learn features from data. Understanding feature learning from a theoretical viewpoint is an emerging field with many open questions. In this work we capture…

无序系统与神经网络 · 物理学 2024-05-20 Kirsten Fischer , Javed Lindner , David Dahmen , Zohar Ringel , Michael Krämer , Moritz Helias

A recent line of work has analyzed the theoretical properties of deep neural networks via the Neural Tangent Kernel (NTK). In particular, the smallest eigenvalue of the NTK has been related to the memorization capacity, the global…

机器学习 · 统计学 2022-08-23 Quynh Nguyen , Marco Mondelli , Guido Montufar

Adversarial examples are inevitable on the road of pervasive applications of deep neural networks (DNN). Imperceptible perturbations applied on natural samples can lead DNN-based classifiers to output wrong prediction with fair confidence…

机器学习 · 计算机科学 2020-11-04 Tao Bai , Jinqi Luo , Jun Zhao

Vulnerability to adversarial attacks is one of the principal hurdles to the adoption of deep learning in safety-critical applications. Despite significant efforts, both practical and theoretical, training deep learning models robust to…

机器学习 · 计算机科学 2024-02-29 Luca Bortolussi , Ginevra Carbone , Luca Laurenti , Andrea Patane , Guido Sanguinetti , Matthew Wicker

In this paper, we study the data-dependent convergence and generalization behavior of gradient methods for neural networks with smooth activation. Our first result is a novel bound on the excess risk of deep networks trained by the logistic…

机器学习 · 计算机科学 2024-12-09 Hossein Taheri , Christos Thrampoulidis , Arya Mazumdar

We propose gradient adversarial training, an auxiliary deep learning framework applicable to different machine learning problems. In gradient adversarial training, we leverage a prior belief that in many contexts, simultaneous gradient…

机器学习 · 计算机科学 2018-06-22 Ayan Sinha , Zhao Chen , Vijay Badrinarayanan , Andrew Rabinovich

A quantum neural network (QNN) is a parameterized mapping efficiently implementable on near-term Noisy Intermediate-Scale Quantum (NISQ) computers. It can be used for supervised learning when combined with classical gradient-based…

量子物理 · 物理学 2023-03-28 Xuchen You , Shouvanik Chakrabarti , Boyang Chen , Xiaodi Wu

State-of-the-art neural network training methods depend on the gradient of the network function. Therefore, they cannot be applied to networks whose activation functions do not have useful derivatives, such as binary and discrete-time…

机器学习 · 统计学 2025-06-24 Luke Eilers , Raoul-Martin Memmesheimer , Sven Goedeke

Deep Neural Network (DNN) are vulnerable to adversarial attacks. As a countermeasure, adversarial training aims to achieve robustness based on the min-max optimization problem and it has shown to be one of the most effective defense…

计算机视觉与模式识别 · 计算机科学 2022-05-03 Yaxin Li , Xiaorui Liu , Han Xu , Wentao Wang , Jiliang Tang

The fact that deep neural networks are susceptible to crafted perturbations severely impacts the use of deep learning in certain domains of application. Among many developed defense models against such attacks, adversarial training emerges…

机器学习 · 计算机科学 2020-07-13 Anh Bui , Trung Le , He Zhao , Paul Montague , Olivier deVel , Tamas Abraham , Dinh Phung

Modern deep learning requires large volumes of data, which could contain sensitive or private information that cannot be leaked. Recent work has shown for homogeneous neural networks a large portion of this training data could be…

机器学习 · 计算机科学 2023-11-13 Noel Loo , Ramin Hasani , Mathias Lechner , Alexander Amini , Daniela Rus

We study the eigenvalue distributions of the Conjugate Kernel and Neural Tangent Kernel associated to multi-layer feedforward neural networks. In an asymptotic regime where network width is increasing linearly in sample size, under random…

机器学习 · 统计学 2020-10-13 Zhou Fan , Zhichao Wang

Catastrophic forgetting is one of the fundamental issues of continual learning because neural networks forget the tasks learned previously when trained on new tasks. The proposed framework is a new path-coordinated framework of continual…

机器学习 · 计算机科学 2025-11-05 Rathin Chandra Shit

Convolutional neural networks (CNNs) have achieved beyond human-level accuracy in the image classification task and are widely deployed in real-world environments. However, CNNs show vulnerability to adversarial perturbations that are…

计算机视觉与模式识别 · 计算机科学 2021-03-09 Desheng Wang , Weidong Jin , Yunpu Wu , Aamir Khan

We identify fragile and robust neurons of deep learning architectures using nodal dropouts of the first convolutional layer. Using an adversarial targeting algorithm, we correlate these neurons with the distribution of adversarial attacks…

机器学习 · 计算机科学 2022-02-01 Chandresh Pravin , Ivan Martino , Giuseppe Nicosia , Varun Ojha

A longstanding goal in the theory of deep learning is to characterize the conditions under which a given neural network architecture will be trainable, and if so, how well it might generalize to unseen data. In this work, we provide such a…

机器学习 · 计算机科学 2020-07-14 Lechao Xiao , Jeffrey Pennington , Samuel S. Schoenholz

Deep Neural Networks (DNN) have been shown to be vulnerable to adversarial examples. Adversarial training (AT) is a popular and effective strategy to defend against adversarial attacks. Recent works (Benz et al., 2020; Xu et al., 2021; Tian…

机器学习 · 计算机科学 2023-02-09 Boqi Li , Weiwei Liu

Transformers have become the dominant architecture in modern machine learning, yet the theoretical understanding of their training dynamics remains limited. This paper develops a rigorous mathematical framework for analyzing gradient-based…

最优化与控制 · 数学 2026-05-19 Raphaël Barboni , Maarten V. de Hoop , Takashi Furuya , Gabriel Peyré