Affiliation:
1. School of Computer Science and Engineering, Nanyang Technological University, Singapore
2. Institute for AI Industry Research, Tsinghua University, Beijing, China
3. School of Software, Shandong University, Shandong, China
Abstract
Federated Learning (FL) bridges the gap between collaborative machine learning and preserving data privacy. To sustain the long-term operation of an FL ecosystem, it is important to attract high-quality data owners with appropriate incentive schemes. As an important building block of such incentive schemes, it is essential to fairly evaluate participants’ contribution to the performance of the final FL model without exposing their private data. Shapley Value (SV)–based techniques have been widely adopted to provide a fair evaluation of FL participant contributions. However, existing approaches incur significant computation costs, making them difficult to apply in practice. In this article, we propose the Guided Truncation Gradient Shapley (GTG-Shapley) approach to address this challenge. It reconstructs FL models from gradient updates for SV calculation instead of repeatedly training with different combinations of FL participants. In addition, we design a guided Monte Carlo sampling approach combined with within-round and between-round truncation to further reduce the number of model reconstructions and evaluations required. This is accomplished through extensive experiments under diverse realistic data distribution settings. The results demonstrate that GTG-Shapley can closely approximate actual Shapley values while significantly increasing computational efficiency compared with the state-of-the-art, especially under non-i.i.d. settings.
Funder
National Research Foundation, Singapore
Joint NTU-WeBank Research Centre on Fintech
Nanyang Technological University, Singapore
RIE 2020 Advanced Manufacturing and Engineering (AME) Programmatic
Joint SDU-NTU Research Centre on Artificial Intelligence (C-FAIR), Shandong University, China
NSFC
SDNSFC
Shandong Provincial Key Research and Development Program
Publisher
Association for Computing Machinery (ACM)
Subject
Artificial Intelligence,Theoretical Computer Science
Reference34 articles.
1. Dynamics of Profit-Sharing Games
2. Polynomial calculation of the Shapley value based on sampling
3. Yiqiang Chen, Xiaodong Yang, Xin Qin, Han Yu, Biao Chen, and Zhiqi Shen. 2020. Dealing with label quality disparity in federated learning. In Federated Learning: Privacy and Incentives, Qiang Yang, Lixin Fan, and Han Yu (Eds.). Springer, 106–120.
4. Ningning Ding, Zhixuan Fang, and Jianwei Huang. 2020. Incentive mechanism design for federated learning with multi-dimensional private information. In WiOPT. 1–8.
5. Joint Service Pricing and Cooperative Relay Communication for Federated Learning
Cited by
26 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献