Machine unlearning is a critical area of research aimed at safeguarding data privacy by enabling the removal of sensitive information from machine learning models. One unique challenge in this field is catastrophic unlearning, where erasing specific data from a well-trained model unintentionally removes essential knowledge, causing the model to deviate significantly from a retrained one. To address this, we introduce a novel approach that regularizes the unlearning process by utilizing synthesized mixup samples, which simulate the data susceptible to catastrophic effects. At the core of our approach is a generator-unlearner framework, MixUnlearn, where a generator adversarially produces challenging mixup examples, and the unlearner effectively forgets target information based on these synthesized data. Specifically, we first introduce a novel contrastive objective to train the generator in an adversarial direction: generating examples that prompt the unlearner to reveal information that should be forgotten, while losing essential knowledge. Then the unlearner, guided by two other contrastive loss terms, processes the synthesized and real data jointly to ensure accurate unlearning without losing critical knowledge, overcoming catastrophic effects. Extensive evaluations across benchmark datasets demonstrate that our method significantly outperforms state-of-the-art approaches, offering a robust solution to machine unlearning. This work not only deepens understanding of unlearning mechanisms but also lays the foundation for effective machine unlearning with mixup augmentation.

The United Nations has urged a moratorium on artificial intelligence systems, such as face scanning and social credit systems, that could be a threat to human rights. Michelle Bachelet, the high commissioner for human rights, said countries should ban AI applications that do not comply with international law. Applications that should be prohibited include government "social scoring" systems that judge people based on their behaviour and certain AI-based tools that categorize people into clusters such as by ethnicity or gender. AI-based technologies can be a force for good but they can also "have negative, even catastrophic, effects if they are used without sufficient regard to how they affect people's human rights," Bachelet said in a statement. Her comments came with a new UN report that examines how countries and businesses have rushed into applying AI systems that affect people's lives and livelihoods without setting up proper safeguards to prevent discrimination and other harms.