Exploiting Supervised Poison Vulnerability to Strengthen Self-Supervised Defense
Availability poisons exploit supervised learning (SL) algorithms by introducing class-related shortcut features in images such that models trained on poisoned data are useless for real-world datasets. Self-supervised learning (SSL), which utilizes augmentations to learn instance discrimination, is regarded as a strong defense against poisoned data. However, by extending the study of SSL across multiple poisons on the CIFAR-10 and ImageNet-100 datasets, we demonstrate that it often performs poorly, far below that of training on clean data. Leveraging the vulnerability of SL to poison attacks, we introduce adversarial training (AT) on SL to obfuscate poison features and guide robust feature learning for SSL. Our proposed defense, designated VESPR (Vulnerability Exploitation of Supervised Poisoning for Robust SSL), surpasses the performance of six previous defenses across seven popular availability poisons. VESPR displays superior performance over all previous defenses, boosting the minimum and average ImageNet-100 test accuracies of poisoned models by 16% and 9%, respectively. Through analysis and ablation studies, we elucidate the mechanisms by which VESPR learns robust class features.
Code (1)
Tasks
Self-Supervised LearningSimilar Papers 제목 키워드 기반
Poisoning Semi-supervised Federated Learning via Unlabeled Data: Attacks and Defenses
Semi-supervised Federated Learning (SSFL) has recently drawn much attention due to its practical consideration, i.e., the clients may only have unlabeled data. In practice, these SSFL systems implement semi-supervised tr…
Federated LearningQuantizationBackdoor Attacks on Speech Emotion Recognition via TTS-Generated Poisoning
Speech Emotion Recognition (SER) systems increasingly leverage self-supervised acoustic representations, yet their vulnerability to training-time attacks remains largely underexplored. This paper presents the first syste…
Speech Emotion RecognitionVulnerability-Aware Poisoning Mechanism for Online RL with Unknown Dynamics
Poisoning attacks on Reinforcement Learning (RL) systems could take advantage of RL algorithm's vulnerabilities and cause failure of the learning. However, prior works on poisoning RL usually either unrealistically assum…
Reinforcement Learning (RL)Disabling Self-Correction in Retrieval-Augmented Generation via Stealthy Retriever Poisoning
Retrieval-Augmented Generation (RAG) has become a standard approach for improving the reliability of large language models (LLMs). Prior work demonstrates the vulnerability of RAG systems by misleading them into generati…
Backdooring Self-Supervised Contrastive Learning by Noisy Alignment
Self-supervised contrastive learning (CL) effectively learns transferable representations from unlabeled data containing images or image-text pairs but suffers vulnerability to data poisoning backdoor attacks (DPCLs). An…
Contrastive Learning