logo
科言猫学术研究的AI总结
language
手机端手机端
  • Phone Feature微信小程序
简体中文简体中文
notice消息通知
博客博客
delete
arrow
返回

Matrix Commitment-Based Ownership Verification for Distributed Machine Learning

delete2026-04-27
delete0
PRE
AI
T
Tianxiu Xie
盖
盖珂珂 (Keke Gai)
J
Jing Yu
祝
祝烈煌 (Liehuang Zhu)
Q
Qi Wu
DOI：10.1109/tpami.2026.3687640delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
分布式机器学习（DML）实现了并行训练，但也引发了模型所有权确定的冲突，因为验证个体客户端在权重更新中的具体贡献仍具挑战性。为解决所有权争议并防止恶意客户端伪造数据，验证各客户端独特训练历史的完整性和有效性，可公平确立其对最终DML模型的局部所有权。本文首次提出了一种新颖的基于矩阵承诺的DML所有权验证（MAMMON）方案，采用简洁的证明确保在有限的计算成本内实现计算完整性与正确性。与计算密集型的SNARK基方法不同，MAMMON构建多线性树结构以降低证明更新成本，并通过标准化训练过程避免复杂算术电路计算。此外，利用客户端身份私钥对权重证明进行水印标记，以保护承诺免遭篡改或未授权获取。大量实验表明，MAMMON在保持DML训练计算完整性方面表现优越，涵盖理论分析与与其他现有基线的比较。
Keyword:
Matrix commitment
ownership verification
distributed machine learning
intellectual property protection

期刊

IEEE Transactions on Pattern Analysis and Machine Intelligence 封面图
IEEE Transactions on Pattern Analysis and Machine Intelligence
IF:
18.6
论文数:
1.1K
被引数:
9.8W

机构

B
beijing institute of technology
学者数:
5.5W
论文数: 4.0W
被引数: 63
T
the university of adelaide
学者数:
328
论文数: 183
被引数: 0
M
Minzu University of China
学者数:
3.3K
论文数: 1.9K
被引数: 6.7K
学者 查看更多机构
引用论文

引用论文

Matproofs
errProceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security
IF0
err2022-11-07
err0
errOAAI
errJing Liu; Liang Feng Zhang
err分享
err收藏
RAI2: Responsible Identity Audit Governing the Artificial Intelligence
errProceedings 2023 Network and Distributed System Security Symposium
IF0
err2023-01-01
err0
errOAAI
errTian Dong; Shaofeng Li; Guoxing Chen; Minhui Xue; Haojin Zhu; Zhen Liu
err分享
err收藏
Dynamic Pricing and Placing for Distributed Machine Learning Jobs: An Online Learning Approach
errIEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
IF17.2
err2023-04-01
err3
PREAI
errZhou, Ruiting; Zhang, Xueying; Lui, John C. S.; Li, Zongpeng
err分享
err收藏
FedGMark: Certifiably Robust Watermarking for Federated Graph LearningFedGMark：面向联邦图学习的可验证鲁棒水印技术
errAdvances in Neural Information Processing Systems 37
IF0
err2024-01-01
err0
PREAI
errYang,Yuxin; Li,Qiang; Hong,Yuan; Wang,Binghui
err分享
err收藏
Experimenting with Zero-Knowledge Proofs of Training
errProceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security
IF0
err2023-11-21
err0
PREAI
errSanjam Garg; Aarushi Goel; Somesh Jha; Saeed Mahloujifar; Mohammad Mahmoody; Guru-Vamsi Policharla; Mingyuan Wang
err分享
err收藏
Proof-of-Learning: Definitions and Practice
err2021 IEEE Symposium on Security and Privacy (SP)
IF0
err2021-05-01
err0
errOAAI
errHengrui Jia; Mohammad Yaghini; Christopher A. Choquette-Choo; Natalie Dullerud; Anvith Thudi; Varun Chandrasekaran; Nicolas Papernot
err分享
err收藏
EIFFeL
errProceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security
IF0
err2022-11-07
err0
PREAI
errAmrita Roy Chowdhury; Chuan Guo; Somesh Jha; Laurens van der Maaten
err分享
err收藏
VeriFL: Communication-Efficient and Fast Verifiable Aggregation for Federated LearningVeriFL: 通信-用于联邦学习的高效且快速的可验证聚合
errIEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
IF8
err2021-01-01
err153
errOAAI
errGuo, Xiaojie; Liu, Zheli; Li, Jin; Gao, Jiqiang; Hou, Boyu; Dong, Changyu; Baker, Thar
err分享
err收藏
Leia: A Lightweight Cryptographic Neural Network Inference System at the Edge
errIEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
IF8
err2022-01-01
err20
PREAI
errLiu, Xiaoning; Wu, Bang; Yuan, Xingliang; Yi, Xun
err分享
err收藏
Proof-of-Learning is Currently More Broken Than You Think
err2023 IEEE 8th European Symposium on Security and Privacy (EuroS&P)
IF0
err2023-07-01
err0
errOAAI
errCongyu Fang; Hengrui Jia; Anvith Thudi; Mohammad Yaghini; Christopher A. Choquette-Choo; Natalie Dullerud; Varun Chandrasekaran; Nicolas Papernot
err分享
err收藏
学者 查看更多内容