paper-with-me

Papers

Bypassing Backdoor Detection Algorithms in Deep Learning

2019-05-31 · Te Juin Lester Tan, Reza Shokri

Deep learning models are vulnerable to various adversarial manipulations of their training data, parameters, and input sample. In particular, an adversary can modify the training data and model parameters to embed backdoors into the model, so the model behaves according to the adversary's objective if the input contains the backdoor features, referred to as the backdoor trigger (e.g., a stamp on an image). The poisoned model's behavior on clean data, however, remains unchanged. Many detection algorithms are designed to detect backdoors on input samples or model parameters, through the statistical difference between the latent representations of adversarial and clean input samples in the poisoned model. In this paper, we design an adversarial backdoor embedding algorithm that can bypass the existing detection algorithms including the state-of-the-art techniques. We design an adaptive adversarial training algorithm that optimizes the original loss function of the model, and also maximizes the indistinguishability of the hidden representations of poisoned data and clean data. This work calls for designing adversary-aware defense mechanisms for backdoor detection.

📄 PDF Abstract BibTeX arXiv:1905.13409

Code (2)

aisafety-hkust/backdoor_safety_tuning pytorch
aisafety-hkust/stable_backdoor_purification pytorch

Tasks

Deep Learning

Similar Papers 제목 키워드 기반

BaFFLe: Backdoor detection via Feedback-based Federated Learning

2020-11-04 · Sebastien Andreina, Giorgia Azzurra Marson, Helen Möllering, Ghassan Karame

Recent studies have shown that federated learning (FL) is vulnerable to poisoning attacks that inject a backdoor into the global model. These attacks are effective even when performed by a single client, and undetectable…

Federated LearningModel Poisoning

Gungnir: Exploiting Stylistic Features in Images for Backdoor Attacks on Diffusion Models

2025-02-28 · Yu Pan, Jiahao Chen, Bingrong Dai, Lin Wang 외

In recent years, Diffusion Models (DMs) have demonstrated significant advances in the field of image generation. However, according to current research, DMs are vulnerable to backdoor attacks, which allow attackers to co…

Backdoor Attackbackdoor defenseImage Generation

Towards Invisible Backdoor Attack on Text-to-Image Diffusion Model

2025-03-22 · Jie Zhang, Zhongqi Wang, Shiguang Shan, Xilin Chen

Backdoor attacks targeting text-to-image diffusion models have advanced rapidly, enabling attackers to implant malicious triggers into these models to manipulate their outputs. However, current backdoor samples often exh…

Backdoor Attack

PiDAn: A Coherence Optimization Approach for Backdoor Attack Detection and Mitigation in Deep Neural Networks

2022-03-17 · Yue Wang, Wenqing Li, Esha Sarkar, Muhammad Shafique 외

Backdoor attacks impose a new threat in Deep Neural Networks (DNNs), where a backdoor is inserted into the neural network by poisoning the training dataset, misclassifying inputs that contain the adversary trigger. The m…

Anomaly DetectionBackdoor Attack

Exploiting Layer-Specific Vulnerabilities to Backdoor Attack in Federated Learning

2026-02-16 · Mohammad Hadi Foroughi, Seyed Hamed Rastegar, Mohammad Sabokrou, Ahmad Khonsari arxiv

Federated learning (FL) enables distributed model training across edge devices while preserving data locality. This decentralized approach has emerged as a promising solution for collaborative learning on sensitive user …

Federated Learning