ML Security PapersLearning to Generate Cross-Task Unexploitable Examples
Haoxuan Qu, Qiuchi Xiang, Yujun Cai et al. · Lancaster University · The University of Queensland +2 more
Defends personal images from unauthorized ML training by generating cross-task imperceptible perturbations that make training data unlearnable across diverse vision tasks
Unexploitable example generation aims to transform personal images into their unexploitable (unlearnable) versions before they are uploaded online, thereby preventing unauthorized exploitation of online personal images. Recently, this task has garnered significant research attention due to its critical relevance to personal data privacy. Yet, despite recent progress, existing methods for this task can still suffer from limited practical applicability, as they can fail to generate examples that are broadly unexploitable across different real-world computer vision tasks. To deal with this problem, in this work, we propose a novel Meta Cross-Task Unexploitable Example Generation (MCT-UEG) framework. At the core of our framework, to optimize the unexploitable example generator for effectively producing broadly unexploitable examples, we design a flat-minima-oriented meta training and testing scheme. Extensive experiments show the efficacy of our framework.