通过正交化卷积核实现高效可控的联邦学习数据遗忘
Orthogonal Soft Pruning for Efficient Class Unlearning
- 用正交卷积核和激活驱动的软剪枝，解耦特征表示
- 单次操作即可遗忘98%以上目标类别，保留97%以上准确率
- 适合需要快速擦除数据的隐私敏感场景
联邦学习中的高效且可控的数据遗忘仍具挑战，尤其在非独立同分布（non-IID）设置下，深层特征纠缠加剧了遗忘与保留之间的权衡。为此，我们提出FedOrtho框架，结合正交化深度卷积核与激活驱动的单次软剪枝（OSP）机制。该方法通过强制核正交性与局部-全局对齐，解耦特征表示并缓解客户端漂移。结构上的独立性使能精准剪除与遗忘相关的卷积核，同时保留需保留的知识。FedOrtho在CIFAR-10、CIFAR100和TinyImageNet上使用ResNet和VGG架构均达到最优性能，支持类别级、客户端级和样本级遗忘，遗忘质量超98%。在联邦设置中计算与通信开销降低2-3个数量级，中心化场景下擦除时间低于1秒，同时保持超过97%的保留准确率并降低成员推断风险。
原文摘要 · Abstract (English)
Efficient and controllable data unlearning in federated learning remains challenging, due to the trade-off between forgetting and retention performance. Especially under non-independent and identically distributed (non-IID) settings, where deep feature entanglement exacerbates this dilemma. To address this challenge, we propose FedOrtho, a federated unlearning framework that combines orthogonalized deep convolutional kernels with an activation-driven controllable one-shot soft pruning (OSP) mechanism. FedOrtho enforces kernel orthogonality and local-global alignment to decouple feature representations and mitigate client drift. This structural independence enables precise one-shot pruning of forgetting-related kernels while preserving retained knowledge. FedOrtho achieves SOTA performance on CIFAR-10, CIFAR100 and TinyImageNet with ResNet and VGG frameworks, verifying that FedOrtho supports class-, client-, and sample-level unlearning with over 98% forgetting quality. It reduces computational and communication costs by 2-3 orders of magnitude in federated settings and achieves subsecond-level erasure in centralized scenarios while maintaining over 97% retention accuracy and mitigating membership inference risks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。