Zero-Knowledge Proof-based Verifiable Decentralized Machine Learning in Communication Network - A Comprehensive Survey

Aug 2, 2026

Machine learning has advanced network communication, improving decision-making, user behavior analysis, and fault detection, while communication networks enable large-scale data collection. Centralized learning, however, depends on users providing data, creating major privacy and security risks. Decentralized learning exchanges computation results instead of raw private data but introduces trust and verifiability challenges, especially for validating others’ shared results. This survey reviews Zero-Knowledge Proof-based Verifiable Machine Learning (ZKP-VML), defines it via four algorithms and key security properties, and organizes the research timeline and schemes by security properties to extract common design principles and optimization strategies.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages39

Reading Time1 h 38 min

AccessFree to download

Introduction

Verifiability challenges in decentralized learning

Centralized vs decentralized ML privacy trade-offs

Federated learning as a decentralized framework

Zero-knowledge proofs for verifiability

Zero-Knowledge Proof-based Verifiable Decentralized Machine Learning in Communication Network: A Comprehensive Survey Zhibo Xing, Zijian Zhang, Member, IEEE, Ziang Zhang, Zhen Li, Meng Li, Senior Member, IEEE, Jiamou Liu, Zongyang Zhang, Yi Zhao, Qi Sun, Liehuang Zhu, Senior Member, IEEE, Giovanni Russello, Member, IEEE. arXiv :2310 . 14848v2 [ cs .LG] 5 Mar 2025 Abstract—Over recent decades, machine learning has significantly advanced network communication, enabling improved decision-making, user behavior analysis, and fault detection. Simultaneously, the growth of communication networks has facilitated the efficient collection of large-scale training data. Traditional centralized machine learning, however, requires data collection from users, raising significant concerns about privacy and security. Decentralized approaches, where participants exchange computation results instead of raw private data, mitigate these risks but introduce challenges related to trust and verifiability. A critical issue arises: How can one ensure the integrity and validity of computation results shared by other participants? Existing survey articles predominantly address security and privacy concerns in decentralized machine learning, whereas this survey uniquely highlights the emerging issue of verifiability. Recognizing the critical role of zero-knowledge proofs in ensuring verifiability, we present a comprehensive review of Zero-Knowledge Proofbased Verifiable Machine Learning (ZKP-VML). To clarify the research problem, we present a definition of ZKP-VML consisting of four algorithms, along with several corresponding key security properties. Besides, we provide an overview of the current research landscape by systematically organizing the research timeline and categorizing existing schemes based on their security properties. Furthermore, through an in-depth analysis of each existing scheme, we summarize their technical contributions and optimization strategies, aiming to uncover common design principles underlying ZKP-VML schemes. Building on the reviews and Zhibo Xing is with the School of Cyberspace Science and Technology, Beijing Institute of Technology, Beijing, 100081, China, and the School of Computer Science, The University of Auckland, Auckland, 1010, New Zealand. E-mail: [3120215670@bit.edu.cn](3120215670@bit.edu.cn). Zijian Zhang is with the School of Cyberspace Science and Technology, Beijing Institute of Technology, Beijing, 100081, China, and Southeast Institute of Information Technology, Beijing Institute of Technology, Fujian, 351100, China. E-mail: [zhangzijian@bit.edu.cn](zhangzijian@bit.edu.cn). Jiamou Liu and Giovanni Russello are with the School of Computer Science, The University of Auckland, Auckland, 1010, New Zealand. Email:{jiamou.liu, [g.russello](g.russello}@auckland.ac.nz)[}](g.russello}@auckland.ac.nz)[@auckland.ac.nz](g.russello}@auckland.ac.nz). Ziang Zhang, Zhen Li, Yi Zhao, Liehuang Zhu are with the School of Cyberspace Science and Technology, Beijing Institute of Technology, Beijing, 100081, China. E-mail: {3220231794, [zhen.li](zhen.li), zhaoyi, [liehuangz](liehuangz}@bit.edu.cn)[}](liehuangz}@bit.edu.cn)[@bit.edu.cn](liehuangz}@bit.edu.cn). Meng Li is with Key Laboratory of Knowledge Engineering with Big Data (Hefei University of Technology), Ministry of Education; School of Computer Science and Information Engineering, Hefei University of Technology, Anhui, 230601, China; Anhui Province Key Laboratory of Industry Safety and Emergency Technology; and Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology) . Email: [mengli@hfut.edu.cn](mengli@hfut.edu.cn). Zongyang Zhang is with the School of Cyber Science and Technology, Beihang University, Beijing, 100191, China. E-mail: [zongyangzhang@buaa.edu.cn](zongyangzhang@buaa.edu.cn). Qi Sun is with the Department of Bioinformatics, Hangzhou Nuowei Information Technology Co.,Ltd, Zhejiang, 310053, China. E-m

FAQs

Why does decentralized machine learning need verifiability?

Decentralized learning shares computation results instead of raw private data, but this creates trust concerns. Verifiability ensures the integrity and validity of results provided by other participants.

What is ZKP-VML and how is it characterized in the survey?

ZKP-VML is defined as a verifiable decentralized machine learning approach empowered by zero-knowledge proofs. The survey presents a definition based on four algorithms and several corresponding security properties.

How does the survey structure the research landscape?

It organizes the research timeline systematically and categorizes existing schemes according to their security properties, then analyzes each scheme’s technical contributions and optimization strategies.

Aug 2, 2026

Upload by Olivia Brown