ToFU: Transforming How Federated Learning Systems Forget User Data
This addresses privacy risks in federated learning systems for users and organizations needing compliance with regulations, though it is incremental as it builds on existing federated unlearning methods.
The paper tackles the problem of neural networks memorizing training data in federated learning, which poses privacy risks, by proposing ToFU, a framework that incorporates transformations during learning to reduce memorization and improve unlearning efficiency. Results show it outperforms existing methods on benchmarks like CIFAR-10 and CIFAR-100, reducing unlearning time.
Neural networks unintentionally memorize training data, creating privacy risks in federated learning (FL) systems, such as inference and reconstruction attacks on sensitive data. To mitigate these risks and to comply with privacy regulations, Federated Unlearning (FU) has been introduced to enable participants in FL systems to remove their data's influence from the global model. However, current FU methods primarily act post-hoc, struggling to efficiently erase information deeply memorized by neural networks. We argue that effective unlearning necessitates a paradigm shift: designing FL systems inherently amenable to forgetting. To this end, we propose a learning-to-unlearn Transformation-guided Federated Unlearning (ToFU) framework that incorporates transformations during the learning process to reduce memorization of specific instances. Our theoretical analysis reveals how transformation composition provably bounds instance-specific information, directly simplifying subsequent unlearning. Crucially, ToFU can work as a plug-and-play framework that improves the performance of existing FU methods. Experiments on CIFAR-10, CIFAR-100, and the MUFAC benchmark show that ToFU outperforms existing FU baselines, enhances performance when integrated with current methods, and reduces unlearning time.