paper-with-me

Papers

Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective

2023-11-28 · Ming-Yu Chung, Sheng-Yen Chou, Chia-Mu Yu, Pin-Yu Chen, Sy-Yen Kuo, Tsung-Yi Ho

Dataset distillation offers a potential means to enhance data efficiency in deep learning. Recent studies have shown its ability to counteract backdoor risks present in original training samples. In this study, we delve into the theoretical aspects of backdoor attacks and dataset distillation based on kernel methods. We introduce two new theory-driven trigger pattern generation methods specialized for dataset distillation. Following a comprehensive set of analyses and experiments, we show that our optimization-based trigger design framework informs effective backdoor attacks on dataset distillation. Notably, datasets poisoned by our designed trigger prove resilient against conventional backdoor attack detection and mitigation methods. Our empirical results validate that the triggers developed using our approaches are proficient at executing resilient backdoor attacks.

📄 PDF Abstract BibTeX arXiv:2311.16646

Code (0)

등록된 구현이 없습니다.

Tasks

Backdoor AttackDataset Distillation

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Risk of Text Backdoor Attacks Under Dataset Distillation

2024-10-17 · Information Security Conference 2024 10 · Kejun Zhang, Yutuo Song, Shaofei Xu, Pengcheng Li 외

Dataset distillation aims to transfer knowledge from large training datasets into smaller datasets to enable rapid training of neural networks while maintaining the original dataset’s performance. However, current resear…

Backdoor AttackDataset Distillationtext-classificationText Classification

INK: Inheritable Natural Backdoor Attack Against Model Distillation

2023-04-21 · Xiaolei Liu, Ming Yi, Kangyi Ding, Bangzhou Xin 외

Deep learning models are vulnerable to backdoor attacks, where attackers inject malicious behavior through data poisoning and later exploit triggers to manipulate deployed models. To improve the stealth and effectiveness…

Backdoor AttackData PoisoningImage Augmentationmodel

Backdoor Attacks Against Dataset Distillation

2023-01-03 · Yugeng Liu, Zheng Li, Michael Backes, Yun Shen 외

Dataset distillation has emerged as a prominent technique to improve data efficiency when training machine learning models. It encapsulates the knowledge from a large dataset into a smaller synthetic dataset. A model tra…

Backdoor AttackDataset Distillation

DHBE: Data-free Holistic Backdoor Erasing in Deep Neural Networks via Restricted Adversarial Distillation

2023-06-13 · Zhicong Yan, Shenghong Li, Ruijie Zhao, Yuan Tian 외

Backdoor attacks have emerged as an urgent threat to Deep Neural Networks (DNNs), where victim DNNs are furtively implanted with malicious neurons that could be triggered by the adversary. To defend against backdoor atta…

Dark Distillation: Backdooring Distilled Datasets without Accessing Raw Data

2025-02-06 · Ziyuan Yang, Ming Yan, Yi Zhang, Joey Tianyi Zhou

Dataset distillation (DD) enhances training efficiency and reduces bandwidth by condensing large datasets into smaller synthetic ones. It enables models to achieve performance comparable to those trained on the raw full …

Dataset Distillation