Zobrazeno 1 - 10
of 299
pro vyhledávání: '"ZHU Mingli"'
Publikováno v:
Open Life Sciences, Vol 17, Iss 1, Pp 447-454 (2022)
This study aimed to examine whether nuclear receptor 4a1 (NR4A1) is involved in inhibiting hepatic stellate cell (HSC) activation and liver fibrosis through the epithelial–mesenchymal transition (EMT). HSC-T6 cells were divided into the control gro
Externí odkaz:
https://doaj.org/article/ff0704c838c1448db124c03b18d48fa3
Autor:
Wu, Baoyuan, Chen, Hongrui, Zhang, Mingda, Zhu, Zihao, Wei, Shaokui, Yuan, Danni, Zhu, Mingli, Wang, Ruotong, Liu, Li, Shen, Chao
As an emerging approach to explore the vulnerability of deep neural networks (DNNs), backdoor learning has attracted increasing interest in recent years, and many seminal backdoor attack and defense algorithms are being developed successively or conc
Externí odkaz:
http://arxiv.org/abs/2407.19845
Deep neural networks face persistent challenges in defending against backdoor attacks, leading to an ongoing battle between attacks and defenses. While existing backdoor defense strategies have shown promising performance on reducing attack success r
Externí odkaz:
http://arxiv.org/abs/2405.16134
Few-shot class-incremental learning (FSCIL) aims to continually fit new classes with limited training data, while maintaining the performance of previously learned classes. The main challenges are overfitting the rare new training samples and forgett
Externí odkaz:
http://arxiv.org/abs/2401.07208
Autor:
Wu, Baoyuan, Wei, Shaokui, Zhu, Mingli, Zheng, Meixi, Zhu, Zihao, Zhang, Mingda, Chen, Hongrui, Yuan, Danni, Liu, Li, Liu, Qingshan
Adversarial phenomenon has been widely observed in machine learning (ML) systems, especially in those using deep neural networks, describing that ML systems may produce inconsistent and incomprehensible predictions with humans at some particular case
Externí odkaz:
http://arxiv.org/abs/2312.08890
Publikováno v:
CVPR 2024
Studying backdoor attacks is valuable for model copyright protection and enhancing defenses. While existing backdoor attacks have successfully infected multimodal contrastive learning models such as CLIP, they can be easily countered by specialized b
Externí odkaz:
http://arxiv.org/abs/2311.12075
Recent studies have demonstrated the susceptibility of deep neural networks to backdoor attacks. Given a backdoored model, its prediction of a poisoned sample with trigger will be dominated by the trigger information, though trigger information and b
Externí odkaz:
http://arxiv.org/abs/2306.16697
Backdoor defense, which aims to detect or mitigate the effect of malicious triggers introduced by attackers, is becoming increasingly critical for machine learning security and integrity. Fine-tuning based on benign data is a natural defense to erase
Externí odkaz:
http://arxiv.org/abs/2304.11823
Publikováno v:
Nonferrous Metals Engineering. Jun2024, Vol. 14 Issue 6, p116-124. 9p.
Development and validation of an HILIC/MS/MS method for determination of nusinersen in rabbit plasma
Autor:
Zhang, Xiao, Sha, Chunjie, Zhang, Wei, Zhao, Fengjuan, Zhu, Mingli, Leng, Guangyi, Liu, Wanhui
Publikováno v:
In Heliyon 30 May 2024 10(10)