提出新方法让量化模型安全删数据，既忘得干净又不伤原有性能。
Forget by Uncertainty: Orthogonal Entropy Unlearning for Quantized Neural Networks
- 用预测不确定性引导遗忘，避免错误记忆到特定类别。
- 通过梯度正交投影，解决遗忘与保留任务的冲突问题。
- 适合需要合规删除数据的边缘设备模型部署场景。
在边缘设备上部署量化神经网络，并结合如GDPR等隐私法规，迫切需要对量化模型进行机器遗忘。然而现有方法存在关键挑战：通过训练模型记忆错误标签来实现遗忘，混淆了遗忘与误记；且采用标量梯度重加权，无法解决遗忘梯度与保留梯度之间的方向冲突。本文提出OEU（Orthogonal Entropy Unlearning）框架，包含两项创新：1）熵引导遗忘通过最大化被遗忘数据上的预测不确定性，提供无偏的遗忘方向，避免向任意特定类别产生自信的错误预测；2）梯度正交投影将遗忘梯度投影至保留梯度的正交补空间，理论上保证了在一阶近似下性能保留。大量实验表明，OEU在遗忘效果和保留准确率方面均优于现有方法。
原文摘要 · Abstract (English)
The deployment of quantized neural networks on edge devices, combined with privacy regulations like GDPR, creates an urgent need for machine unlearning in quantized models. However, existing methods face critical challenges: they induce forgetting by training models to memorize incorrect labels, conflating forgetting with misremembering, and employ scalar gradient reweighting that cannot resolve directional conflicts between gradients. We propose OEU, a novel Orthogonal Entropy Unlearning framework with two key innovations: 1) Entropy-guided unlearning provides an unbiased forgetting direction by maximizing prediction uncertainty on forgotten data, avoiding confident misprediction toward any specific class, and 2) Gradient orthogonal projection eliminates interference by projecting forgetting gradients onto the orthogonal complement of retain gradients, providing theoretical guarantees for utility preservation under first-order approximation. Extensive experiments demonstrate that OEU outperforms existing methods in both forgetting effectiveness and retain accuracy.
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。