Výsledky vyhledávání - "Zhong, Xuyang"

Report

Towards Efficient Training and Evaluation of Robust Models against $l_0$ Bounded Adversarial Perturbations

Autor: Zhong, Xuyang, Huang, Yixiao, Liu, Chen

This work studies sparse adversarial perturbations bounded by $l_0$ norm. We propose a white-box PGD-like attack method named sparse-PGD to effectively and efficiently generate such perturbations. Furthermore, we combine sparse-PGD with a black-box a

Externí odkaz: http://arxiv.org/abs/2405.05075

Zobrazit plný text záznamu

Report

Towards Mitigating Architecture Overfitting in Dataset Distillation

Autor: Zhong, Xuyang, Liu, Chen

Dataset distillation methods have demonstrated remarkable performance for neural networks trained with very limited training data. However, a significant challenge arises in the form of architecture overfitting: the distilled training data synthesize

Externí odkaz: http://arxiv.org/abs/2309.04195

Zobrazit plný text záznamu

Akademický článek

Research on the Algorithm for Height Measurement of Sand Pipe Based on Deep Learning

Autor: YUAN Hongxiang, WANG Junping, SHEN Yunbo, ZHONG Xuyang, JIANG Haixiao

Publikováno v: Kongzhi Yu Xinxi Jishu, Iss 2, Pp 78-84 (2023)

During the daily operation of locomotive, affected by the locomotive itself and external environmental factors, the height of the sand pipe may change, which will lead to wheel slippage failure in severe cases, resulting in safety problems. Therefore

Externí odkaz: https://doaj.org/article/49cbdd4d97c949348a6e68624248c38d