针对联邦时序预测模型，提出新型梯度反演攻击，显著提升隐私泄露风险。
TS-Inverse: A Gradient Inversion Attack Tailored for Federated Time Series Forecasting Models
- 设计量化预测的梯度反演模型，结合周期与趋势正则化。
- 在4个时序数据集上，sMAPE指标提升2至10倍。
- 适用于研究联邦学习隐私漏洞的开发者与安全研究人员。
联邦时序预测（TSF）使具有隐私敏感时序数据的客户端可协同训练精准预测模型，如能源负荷预测。然而，联邦学习中的隐私风险依然存在，服务器可能通过梯度反演攻击（GIA）重构客户端训练数据。尽管已有针对图像分类任务的GIA研究，但对时序回归任务的研究极少。本文首次在4种TSF模型和4个数据集上开展系统性实证研究，揭示了重建时序观测值与目标值的独特挑战。随后提出TS-Inverse，一种新型GIA方法，通过（i）学习输出分位数预测的梯度反演模型，（ii）引入包含周期性和趋势正则化的损失函数，（iii）基于分位数预测进行正则化，显著提升反演效果。实验表明，TS-Inverse在时序数据上的性能相比现有GIA方法，至少实现2至10倍的sMAPE指标改进。代码仓库：https://github.com/Capsar/ts-inverse
原文摘要 · Abstract (English)
Federated learning (FL) for time series forecasting (TSF) enables clients with privacy-sensitive time series (TS) data to collaboratively learn accurate forecasting models, for example, in energy load prediction. Unfortunately, privacy risks in FL persist, as servers can potentially reconstruct clients' training data through gradient inversion attacks (GIA). Although GIA is demonstrated for image classification tasks, little is known about time series regression tasks. In this paper, we first conduct an extensive empirical study on inverting TS data across 4 TSF models and 4 datasets, identifying the unique challenges of reconstructing both observations and targets of TS data. We then propose TS-Inverse, a novel GIA that improves the inversion of TS data by (i) learning a gradient inversion model that outputs quantile predictions, (ii) a unique loss function that incorporates periodicity and trend regularization, and (iii) regularization according to the quantile predictions. Our evaluations demonstrate a remarkable performance of TS-Inverse, achieving at least a 2x-10x improvement in terms of the sMAPE metric over existing GIA methods on TS data. Code repository: https://github.com/Capsar/ts-inverse
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。