Results 51 to 60 of about 29,955 (264)

Image Classification Adversarial Example Defense Method Based on Conditional Diffusion Model [PDF]

Jisuanji gongcheng
Deep-learning models have achieved impressive results in fields such as image classification; however, they remain vulnerable to interference and threats from adversarial examples.
CHEN Zimin, GUAN Zhitao
doaj   +1 more source

Adversarial Examples in Physical World [PDF]

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, 2021
Although deep neural networks (DNNs) have already made fairly high achievements and a very wide range of impact, their vulnerability attracts lots of interest of researchers towards related studies about artificial intelligence (AI) safety and robustness this year.
openaire   +1 more source

AT‐AER: Adversarial Training With Adaptive Example Reuse

CAAI Transactions on Intelligence Technology
Adversarial training (AT) is widely regarded as a crucial defense method for deep neural networks against adversarial attacks. Most of the existing AT methods suffer from the problems of insufficient coverage of perturbation space and robust overfitting.
Meng Hu, Yanting Guo, Ran Wang, Xizhao Wang, Rihao Li, Qin Wang   +5 more
doaj   +1 more source

A survey of practical adversarial example attacks

Cybersecurity, 2018
Adversarial examples revealed the weakness of machine learning techniques in terms of robustness, which moreover inspired adversaries to make use of the weakness to attack systems employing machine learning.
Lu Sun, Mingtian Tan, Zhe Zhou
doaj   +1 more source

DTFA: Adversarial attack with discrete cosine transform noise and target features on deep neural networks

IET Image Processing, 2023
Image recognition on deep neural network is vulnerable to adversarial sample attacks. The adversarial attack accuracy is low when only limited queries on the target are allowed with the current black box environment.
Dong Yang, Wei Chen, Songjie Wei
doaj   +1 more source

When adversarial examples are excusable

CoRR, 2022
Neural networks work remarkably well in practice and theoretically they can be universal approximators. However, they still make mistakes and a specific type of them called adversarial errors seem inexcusable to humans. In this work, we analyze both test errors and adversarial errors on a well controlled but highly non-linear visual classification ...
Pieter-Jan Kindermans, Charles Staats
openaire   +2 more sources

A Multi‐Scale Machine Learning Framework for the Inverse Design of High Entropy Alloys

Advanced Engineering Materials, EarlyView.
High‐entropy alloys offer vast potential for various applications, including electrocatalysis; however, their compositional complexity challenges conventional screening. We introduce an inverse‐design framework combining two neural networks to determine optimal compositions and reconstruct nanoparticle geometry from targeted properties and conventional
Mikael Takoutsin, Marta Campolucci, Nicolas de Andrade Ishiki, Nathan Mirgot, Nadir Zhamantay, Frédéric Kanoufi, Jennifer Peron, Jean Charlety, Emiliano Fonda, Valérie Briois, Marco Faustini, Maria‐Letizia De Marco, Christine Goyhenex, Ovidiu Ersen, Hervé Bulou   +14 more
wiley   +1 more source

DualFlow: Generating imperceptible adversarial examples by flow field and normalize flow-based model

Frontiers in Neurorobotics, 2023
Recent adversarial attack research reveals the vulnerability of learning-based deep learning models (DNN) against well-designed perturbations. However, most existing attack methods have inherent limitations in image quality as they rely on a relatively ...
Renyang Liu, Renyang Liu, Xin Jin, Xin Jin, Dongting Hu, Jinhong Zhang, Jinhong Zhang, Yuanyu Wang, Jin Zhang, Wei Zhou, Wei Zhou   +10 more
doaj   +1 more source

Adversarial Examples Are Not Bugs, They Are Features

CoRR, 2019
Adversarial examples have attracted significant attention in machine learning, but the reasons for their existence and pervasiveness remain unclear. We demonstrate that adversarial examples can be directly attributed to the presence of non-robust features: features derived from patterns in the data distribution that are highly predictive, yet brittle ...
Ilyas, A, Santurkar, S, Tsipras, D, Engstrom, L, Tran, B, Madry, A   +5 more
openaire   +4 more sources

Functionalizing Micro‐to‐Mesoscopic Electrode Architectures for Regulating Electron Transfer Behaviors in Electrocatalysis

Advanced Functional Materials, EarlyView.
A systematic review is conducted to assess the influence of electrode architecture across micro‐ to mesoscopic length scales on electron‐transfer pathways in electrocatalysis. We discuss the structure‐activity relationships in electrocatalytic applications, including resource recovery and environmental remediation, and provide cost‐effective, efficient
Manshu Zhao, Jiaojiao Zhang, Shideng Yuan, Xinmeng Yu, Keming Wan, Xiaoguang Duan, Zhining Wang   +6 more
wiley   +1 more source