Open Access

ARTICLE

MMH-FE: A Multi-Precision and Multi-Sourced Heterogeneous Privacy-Preserving Neural Network Training Based on Functional Encryption

Hao Li^1,#, Kuan Shao^1,#, Xin Wang², Mufeng Wang³, Zhenyong Zhang^1,2,*

1 The State Key Laboratory of Public Big Data, College of Computer Science and Technology, Guizhou University, Guiyang, 550025, China
2 Key Laboratory of Computing Power Network and Information Security, Ministry of Education, Shandong Computer Science Center, Qilu University of Technology (Shandong Academy of Sciences), Jinan, 250014, China
3 China Industrial Control Systems Cyber Emergency Response Team, Beijing, 100040, China

* Corresponding Author: Zhenyong Zhang. Email:
# These authors contributed equally to this work

(This article belongs to the Special Issue: Privacy-Preserving Deep Learning and its Advanced Applications)

Computers, Materials & Continua 2025, 82(3), 5387-5405. https://doi.org/10.32604/cmc.2025.059718

Received 15 October 2024; Accepted 02 January 2025; Issue published 06 March 2025

Abstract

Due to the development of cloud computing and machine learning, users can upload their data to the cloud for machine learning model training. However, dishonest clouds may infer user data, resulting in user data leakage. Previous schemes have achieved secure outsourced computing, but they suffer from low computational accuracy, difficult-to-handle heterogeneous distribution of data from multiple sources, and high computational cost, which result in extremely poor user experience and expensive cloud computing costs. To address the above problems, we propose a multi-precision, multi-sourced, and multi-key outsourcing neural network training scheme. Firstly, we design a multi-precision functional encryption computation based on Euclidean division. Second, we design the outsourcing model training algorithm based on a multi-precision functional encryption with multi-sourced heterogeneity. Finally, we conduct experiments on three datasets. The results indicate that our framework achieves an accuracy improvement of to . Additionally, it offers a memory space optimization of times compared to the previous best approach.

---

Keywords

---