arXiv 2410.068482024-10-09cs.LG

通过变换实现遗忘：借助类别感知表征变换赋能联邦遗忘学习

Forgetting Through Transforming: Enabling Federated Unlearning via Class-Aware Representation Transformation

Shanghai Jiao Tong University（上海交通大学）；National University of Singapore（新加坡国立大学）

Qi Guo, Zhen Tian, Minghao Yao, Yong Qi, Saiyu Qi, Yun Li, Jin Song Dong

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对现有联邦遗忘方法在非独立同分布场景下难以平衡擦除效果与模型效用的问题，提出基于类别感知表征变换的FUCRT方法，通过变换类别选择和双对比对齐实现高效遗忘，性能优于现有基线。

摘要：

联邦遗忘学习（Federated Unlearning, FU）可让客户端选择性地移除特定数据对已训练联邦学习模型的影响，以解决隐私顾虑并满足监管要求。但现有FU方法往往难以在有效擦除与模型效用保持之间取得平衡，非独立同分布（non-IID）场景下的类别级遗忘任务中尤为突出。\n本文提出基于类别感知表征变换的联邦遗忘方法（FUCRT），这是一种通过类别感知表征变换实现遗忘的新方法。FUCRT包含两个核心组件：（1）变换类别选择策略，用于识别最优遗忘方向；（2）变换对齐技术，采用双类别感知对比学习确保不同客户端的变换过程一致。\n在四个数据集上开展的大量实验表明，FUCRT在擦除保证、模型效用保持和效率方面均表现更优。它可实现待遗忘类别100%的完全擦除，同时保持甚至提升剩余类别的性能，在IID和Non-IID两种场景下均优于现有最优基线方法。对表征空间的分析显示，FUCRT能有效将待遗忘类别的表征与剩余类别中的变换类别表征融合，效果高度接近从零重新训练的模型。

英文摘要：

Federated Unlearning (FU) enables clients to selectively remove the influence of specific data from a trained federated learning model, addressing privacy concerns and regulatory requirements. However, existing FU methods often struggle to balance effective erasure with model utility preservation, especially for class-level unlearning in non-IID settings. We propose Federated Unlearning via Class-aware Representation Transformation (FUCRT), a novel method that achieves unlearning through class-aware representation transformation. FUCRT employs two key components: (1) a transformation class selection strategy to identify optimal forgetting directions, and (2) a transformation alignment technique using dual class-aware contrastive learning to ensure consistent transformations across clients. Extensive experiments on four datasets demonstrate FUCRT's superior performance in terms of erasure guarantee, model utility preservation, and efficiency. FUCRT achieves complete (100\%) erasure of unlearning classes while maintaining or improving performance on remaining classes, outperforming state-of-the-art baselines across both IID and Non-IID settings. Analysis of the representation space reveals FUCRT's ability to effectively merge unlearning class representations with the transformation class from remaining classes, closely mimicking the model retrained from scratch.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。