Autoregressive Perturbations for Data PoisoningDownload PDF

Published: 31 Oct 2022, 18:00, Last Modified: 02 Jan 2023, 02:20NeurIPS 2022 AcceptReaders: Everyone
Keywords: autoregressive processes, poisons, data poisoning, data protection, imperceptible perturbations, adversarial machine learning
Abstract: The prevalence of data scraping from social media as a means to obtain datasets has led to growing concerns regarding unauthorized use of data. Data poisoning attacks have been proposed as a bulwark against scraping, as they make data ``unlearnable'' by adding small, imperceptible perturbations. Unfortunately, existing methods require knowledge of both the target architecture and the complete dataset so that a surrogate network can be trained, the parameters of which are used to generate the attack. In this work, we introduce autoregressive (AR) poisoning, a method that can generate poisoned data without access to the broader dataset. The proposed AR perturbations are generic, can be applied across different datasets, and can poison different architectures. Compared to existing unlearnable methods, our AR poisons are more resistant against common defenses such as adversarial training and strong data augmentations. Our analysis further provides insight into what makes an effective data poison.
Supplementary Material: zip
18 Replies