Privacy-Preserving Machine Learning Training in IoT Aggregation Scenarios

In developing smart city, the growing popularity of machine learning (ML) that appreciates high-quality training data sets generated from diverse Internet-of-Things (IoT) devices raises natural questions about the privacy guarantees that can be provided in such settings. Privacy-preserving ML training in an aggregation scenario enables a model demander to securely train ML models with the sensitive IoT data gathered from IoT devices. The existing solutions are generally server aided, cannot deal with the collusion threat between the servers or between the servers and data owners, and do not match the delicate environments of IoT. We propose a privacy-preserving ML training framework named Heda that consists of a library of building blocks based on partial homomorphic encryption, which enables constructing multiple privacy-preserving ML training protocols for the aggregation scenario without the assistance of untrusted servers, and defending the security under collusion situations. Rigorous security analysis demonstrates the proposed protocols can protect the privacy of each participant in the honest-but-curious model and guarantee the security under most collusion situations. Extensive experiments validate the efficiency of Heda, which achieves privacy-preserving ML training without losing the model accuracy.

I. Introduction

Internet of Things (IoT) plays a remarkable role in all aspects of our daily lives, covering various fields, including healthcare, industrial appliances, sports, homes, etc. [1], [2]. The large data collected from IoT devices with machine learning (ML) technologies have been accelerating smart city steps and improving our daily lives [3]–[5]. For a personal healthcare example, fitness records monitored by wearable IoT sensors can be used to train an ML model of medical research institutions, for self-rated health measurement.