Sky of Unlearning (SoUL): Rewiring Federated Machine Unlearning via Selective Pruning
This addresses security risks like data poisoning in resource-constrained drone networks, offering a scalable solution, though it appears incremental as it builds on existing federated unlearning methods.
The paper tackles the problem of removing adversarial data contributions in federated learning for Internet of Drones networks by proposing the Sky of Unlearning (SoUL) framework, which uses selective pruning to efficiently unlearn data while maintaining model accuracy comparable to full retraining and reducing computational and communication overhead.
The Internet of Drones (IoD), where drones collaborate in data collection and analysis, has become essential for applications such as surveillance and environmental monitoring. Federated learning (FL) enables drones to train machine learning models in a decentralized manner while preserving data privacy. However, FL in IoD networks is susceptible to attacks like data poisoning and model inversion. Federated unlearning (FU) mitigates these risks by eliminating adversarial data contributions, preventing their influence on the model. This paper proposes sky of unlearning (SoUL), a federated unlearning framework that efficiently removes the influence of unlearned data while maintaining model performance. A selective pruning algorithm is designed to identify and remove neurons influential in unlearning but minimally impact the overall performance of the model. Simulations demonstrate that SoUL outperforms existing unlearning methods, achieves accuracy comparable to full retraining, and reduces computation and communication overhead, making it a scalable and efficient solution for resource-constrained IoD networks.