BlockFUL: Enabling Unlearning in Blockchained Federated Learning
This addresses the problem of efficient model editing in secure federated learning systems for applications requiring data integrity, though it is incremental as it builds on existing blockchain and unlearning methods.
The paper tackles the challenge of enabling unlearning in blockchain-based federated learning, where complex inheritance relationships complicate model updates, and introduces BlockFUL, a framework that reduces data dependency and operational overhead, with experiments showing performance improvements on CIFAR-10 and Fashion-MNIST datasets using models like AlexNet and ResNet18.
Unlearning in Federated Learning (FL) presents significant challenges, as models grow and evolve with complex inheritance relationships. This complexity is amplified when blockchain is employed to ensure the integrity and traceability of FL, where the need to edit multiple interlinked blockchain records and update all inherited models complicates the process.In this paper, we introduce Blockchained Federated Unlearning (BlockFUL), a novel framework with a dual-chain structure comprising a live chain and an archive chain for enabling unlearning capabilities within Blockchained FL. BlockFUL introduces two new unlearning paradigms, i.e., parallel and sequential paradigms, which can be effectively implemented through gradient-ascent-based and re-training-based unlearning methods. These methods enhance the unlearning process across multiple inherited models by enabling efficient consensus operations and reducing computational costs. Our extensive experiments validate that these methods effectively reduce data dependency and operational overhead, thereby boosting the overall performance of unlearning inherited models within BlockFUL on CIFAR-10 and Fashion-MNIST datasets using AlexNet, ResNet18, and MobileNetV2 models.