paper-with-me

Papers

Clipping free attacks against artificial neural networks

2018-03-26 · Boussad ADDAD, Jerome Kodjabachian, Christophe Meyer

During the last years, a remarkable breakthrough has been made in AI domain thanks to artificial deep neural networks that achieved a great success in many machine learning tasks in computer vision, natural language processing, speech recognition, malware detection and so on. However, they are highly vulnerable to easily crafted adversarial examples. Many investigations have pointed out this fact and different approaches have been proposed to generate attacks while adding a limited perturbation to the original data. The most robust known method so far is the so called C&W attack [1]. Nonetheless, a countermeasure known as feature squeezing coupled with ensemble defense showed that most of these attacks can be destroyed [6]. In this paper, we present a new method we call Centered Initial Attack (CIA) whose advantage is twofold : first, it insures by construction the maximum perturbation to be smaller than a threshold fixed beforehand, without the clipping process that degrades the quality of attacks. Second, it is robust against recently introduced defenses such as feature squeezing, JPEG encoding and even against a voting ensemble of defenses. While its application is not limited to images, we illustrate this using five of the current best classifiers on ImageNet dataset among which two are adversarialy retrained on purpose to be robust against attacks. With a fixed maximum perturbation of only 1.5% on any pixel, around 80% of attacks (targeted) fool the voting ensemble defense and nearly 100% when the perturbation is only 6%. While this shows how it is difficult to defend against CIA attacks, the last section of the paper gives some guidelines to limit their impact.

📄 PDF Abstract BibTeX arXiv:1803.09468

Code (0)

등록된 구현이 없습니다.

Tasks

Malware Detectionspeech-recognitionSpeech Recognition

Similar Papers 제목 키워드 기반

Clipping Free Attacks Against Neural Networks

2018-01-01 · ICLR 2018 1 · Boussad ADDAD

During the last years, a remarkable breakthrough has been made in AI domain thanks to artificial deep neural networks that achieved a great success in many machine learning tasks in computer vision, natural language proc…

Malware Detectionspeech-recognitionSpeech Recognition

Adaptive Gradient Clipping for Robust Federated Learning

2024-05-23 · Youssef Allouah, Rachid Guerraoui, Nirupam Gupta, Ahmed Jellouli 외

Robust federated learning aims to maintain reliable performance despite the presence of adversarial or misbehaving workers. While state-of-the-art (SOTA) robust distributed gradient descent (Robust-DGD) methods were prov…

ARCFederated Learningimage-classificationImage Classification

Local Differential Privacy is Not Enough: A Sample Reconstruction Attack against Federated Learning with Local Differential Privacy

2025-02-12 · Zhichao You, Xuewen Dong, Shujun Li, Ximeng Liu 외

Reconstruction attacks against federated learning (FL) aim to reconstruct users' samples through users' uploaded gradients. Local differential privacy (LDP) is regarded as an effective defense against various attacks, in…

DenoisingFederated LearningReconstruction Attack

Improved Activation Clipping for Universal Backdoor Mitigation and Test-Time Detection

2023-08-08 · Hang Wang, Zhen Xiang, David J. Miller, George Kesidis

Deep neural networks are vulnerable to backdoor attacks (Trojans), where an attacker poisons the training set with backdoor triggers so that the neural network learns to classify test-time triggers to the attacker's desi…

image-classificationImage Classification

Robust Estimation of Sparse Numerical Vectors under Local Differential Privacy

2026-07-30 · Puning Zhao, Zhikun Zhang, Shaowei Wang, Sheng Yue 외 arxiv

Local differential privacy (LDP) protocols are vulnerable to poisoning attacks. Existing research have proposed efficient defense strategies for single-item users. However, in practice, a user may possess multiple items.…