arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2304.00083cs.CRcs.LG
收藏

一种用于机器学习即服务推理低成本结果验证的生成式框架

A Generative Framework for Low-Cost Result Validation of Machine Learning-as-a-Service Inference

Saint Louis University(圣路易斯大学);New Mexico State University(新墨西哥州立大学)

Abhinav Kumar, Miguel A. Guirao Aguilera, Reza Tourani, Satyajayant Misra

AI总结:

针对外包MLaaS推理完整性验证的计算开销问题,提出Fides框架,利用贪婪蒸馏迁移学习在可信执行环境内高效验证服务模型,并结合生成对抗网络训练攻击检测与重分类模型,实现高达98%的攻击检测准确率和94%的重分类准确率。

摘要:

机器学习(ML)日益普及,促使其被部署到各种敏感领域,这也引发了大量针对ML安全与隐私的研究。然而,在增强/虚拟现实等应用中,外包ML任务的完整性验证更为关键——这一方面尚未受到太多关注。现有解决方案(如多方计算和基于证明的系统)会带来巨大的计算开销,使其不适合实时应用。我们提出了Fides,一种用于机器学习即服务(MLaaS)推理实时完整性验证的新颖框架。Fides采用了一种新颖高效的蒸馏技术——贪婪蒸馏迁移学习——该技术动态地蒸馏并微调一个空间和计算高效的验证模型,以便在可信执行环境内运行时验证相应的服务模型。Fides具有客户端攻击检测模型,该模型使用统计分析和散度测量,以极大概率识别服务模型是否受到攻击。Fides还提供了重分类功能,可在识别出攻击时预测原始类别。我们设计了一个生成对抗网络框架来训练攻击检测和重分类模型。评估表明,Fides的攻击检测准确率高达98%,重分类准确率达94%。

英文摘要:

The growing popularity of Machine Learning (ML) has led to its deployment in various sensitive domains, which has resulted in significant research focused on ML security and privacy. However, in some applications, such as Augmented/Virtual Reality, integrity verification of the outsourced ML tasks is more critical--a facet that has not received much attention. Existing solutions, such as multi-party computation and proof-based systems, impose significant computation overhead, which makes them unfit for real-time applications. We propose Fides, a novel framework for real-time integrity validation of ML-as-a-Service (MLaaS) inference. Fides features a novel and efficient distillation technique--Greedy Distillation Transfer Learning--that dynamically distills and fine-tunes a space and compute-efficient verification model for verifying the corresponding service model while running inside a trusted execution environment. Fides features a client-side attack detection model that uses statistical analysis and divergence measurements to identify, with a high likelihood, if the service model is under attack. Fides also offers a re-classification functionality that predicts the original class whenever an attack is identified. We devised a generative adversarial network framework for training the attack detection and re-classification models. The evaluation shows that Fides achieves an accuracy of up to 98% for attack detection and 94% for re-classification.

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。

↑