VeriSplit: Secure and Practical Offloading of Machine Learning Inferences across IoT Devices

TL;DR

VeriSplit enables secure, private, and efficient offloading of machine learning inferences from IoT devices to local hardware, reducing latency without compromising data privacy or model integrity.

Contribution

We introduce VeriSplit, a novel framework that allows privacy-preserving inference offloading to local devices without modifying existing models or hardware.

Findings

Reduces inference latency by up to 83%

Protects data privacy and model confidentiality effectively

Does not require computation over finite fields

Abstract

Many Internet-of-Things (IoT) devices rely on cloud computation resources to perform machine learning inferences. This is expensive and may raise privacy concerns for users. Consumers of these devices often have hardware such as gaming consoles and PCs with graphics accelerators that are capable of performing these computations, which may be left idle for significant periods of time. While this presents a compelling potential alternative to cloud offloading, concerns about the integrity of inferences, the confidentiality of model parameters, and the privacy of users' data mean that device vendors may be hesitant to offload their inferences to a platform managed by another manufacturer. We propose VeriSplit, a framework for offloading machine learning inferences to locally-available devices that address these concerns. We introduce masking techniques to protect data privacy and model confidentiality, and a commitment-based verification protocol to address integrity. Unlike much prior work aimed at addressing these issues, our approach does not rely on computation over finite field elements, which may interfere with floating-point computation supports on hardware accelerators and require modification to existing models. We implemented a prototype of VeriSplit and our evaluation results show that, compared to performing computation locally, our secure and private offloading solution can reduce inference latency by 28%--83%.