Výsledky vyhledávání - "Zhou, Tian Yi"

Report

Optimal Classification-based Anomaly Detection with Neural Networks: Theory and Practice

Autor: Zhou, Tian-Yi, Lau, Matthew, Chen, Jizhou, Lee, Wenke, Huo, Xiaoming

Anomaly detection is an important problem in many application areas, such as network security. Many deep learning methods for unsupervised anomaly detection produce good empirical performance but lack theoretical guarantees. By casting anomaly detect

Externí odkaz: http://arxiv.org/abs/2409.08521

Zobrazit plný text záznamu

Report

Approximation of RKHS Functionals by Neural Networks

Autor: Zhou, Tian-Yi, Suh, Namjoon, Cheng, Guang, Huo, Xiaoming

Motivated by the abundance of functional data such as time series and images, there has been a growing interest in integrating such data into neural networks and learning maps from function spaces to R (i.e., functionals). In this paper, we study the

Externí odkaz: http://arxiv.org/abs/2403.12187

Zobrazit plný text záznamu

Report

Classification of Data Generated by Gaussian Mixture Models Using Deep ReLU Networks

Autor: Zhou, Tian-Yi, Huo, Xiaoming

This paper studies the binary classification of unbounded data from ${\mathbb R}^d$ generated under Gaussian Mixture Models (GMMs) using deep ReLU neural networks. We obtain $\unicode{x2013}$ for the first time $\unicode{x2013}$ non-asymptotic upper

Externí odkaz: http://arxiv.org/abs/2308.08030

Zobrazit plný text záznamu

Report

Learning Ability of Interpolating Deep Convolutional Neural Networks

Autor: Zhou, Tian-Yi, Huo, Xiaoming

It is frequently observed that overparameterized neural networks generalize well. Regarding such phenomena, existing theoretical work mainly devotes to linear settings or fully-connected neural networks. This paper studies the learning ability of an

Externí odkaz: http://arxiv.org/abs/2210.14184

Zobrazit plný text záznamu