提出知识蒸发与浓缩机制，高效清除特定数据记忆而不影响其他数据。
Siamese Machine Unlearning with Knowledge Vaporization and Concentration
- 用孪生网络实现知识选择性删除，无需额外内存和完整数据集。
- 在多种场景下有效移除遗忘数据的知识，提升模型对剩余数据的性能。
- 适合需要合规删除数据且保持模型可用性的实际应用。
针对“被遗忘权”及去除不良数据的实际需求，机器遗忘成为从训练模型中移除部分数据知识的关键技术。然而现有方法存在方法论支持不足、计算复杂度高、内存消耗大等问题。本文提出知识蒸发与浓缩的概念，通过孪生网络实现对特定数据点学习知识的选择性擦除，同时保留其余数据的表征。所提孪生遗忘方法无需额外内存开销，也无需完全访问剩余数据集。在多个遗忘场景下的大量实验表明，该方法优于基线方法，能有效移除遗忘数据的知识，提升模型在剩余数据上的性能，并降低成员推断攻击的敏感性。
原文摘要 · Abstract (English)
In response to the practical demands of the ``right to be forgotten" and the removal of undesired data, machine unlearning emerges as an essential technique to remove the learned knowledge of a fraction of data points from trained models. However, existing methods suffer from limitations such as insufficient methodological support, high computational complexity, and significant memory demands. In this work, we propose the concepts of knowledge vaporization and concentration to selectively erase learned knowledge from specific data points while maintaining representations for the remaining data. Utilizing the Siamese networks, we exemplify the proposed concepts and develop an efficient method for machine unlearning. Our proposed Siamese unlearning method does not require additional memory overhead and full access to the remaining dataset. Extensive experiments conducted across multiple unlearning scenarios showcase the superiority of Siamese unlearning over baseline methods, illustrating its ability to effectively remove knowledge from forgetting data, enhance model utility on remaining data, and reduce susceptibility to membership inference attacks.
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。