Abstract
The Internet of Federated Things (IoFT) represents a network of interconnected systems with federated learning as the backbone, facilitating collaborative knowledge acquisition while ensuring data privacy for individual systems. The wide adoption of IoFT, however, is hindered by security concerns, particularly the susceptibility of federated learning networks to adversarial attacks. In this paper, we propose an effective non-parametric approach FedRR, which leverages the low-rank features of the transmitted parameter updates generated by federated learning to address the adversarial attack problem. Besides, our proposed method is capable of accurately detecting adversarial clients and controlling the false alarm rate under the scenario with no attack occurring. Experiments based on digit recognition using the MNIST datasets validated the advantages of our approach.
Disclaimer
As a service to authors and researchers we are providing this version of an accepted manuscript (AM). Copyediting, typesetting, and review of the resulting proofs will be undertaken on this manuscript before final publication of the Version of Record (VoR). During production and pre-press, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal relate to these versions also.Additional information
Notes on contributors
Xianjian Xie
Xianjian Xie received his BS degree with a major in Mathematics and a minor in Computer Science from the University of Pittsburgh in 2020. He obtained his MS degree in Data Science from the University of Minnesota in 2022. He is currently pursuing a PhD degree in Data Science at Arizona State University. His research interests focus on the monitoring and modeling of the Internet of Federated Things.
Xiaochen Xian
Xiaochen Xian received the B.S. degree in Mathematics from Zhejiang University, Hangzhou, China in 2014, and the M.S. degree in Statistics and the Ph.D. degree in Industrial Engineering from University of Wisconsin-Madison in 2017 and 2019. Currently, she is an assistant professor in the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Institute of Technology. Prior to joining Georgia Tech, she was an assistant professor at the Department of Industrial and Systems Engineering, University of Florida. Her research interests are focused on big data analytics and system informatics applied various types of computationally aware systems.
Dan Li
Dan Li is an Assistant Professor in the Department of Industrial and Systems Engineering in the University of Wisconsin-Madison. Prior to joining UW-Madison, she was an assistant professor in the Department of Industrial Engineering at Clemson University. She received her Ph.D. in Industrial Engineering and M.S. in Statistics from Georgia Institute of Technology in 2021 and 2020, and received her B.S. in Mechanical (Automotive) Engineering from Tsinghua University, Beijing, China, in 2015. Her research interest lies in developing new data-driven algorithms that are tailored for enhancing the cyber-physical resilience and security of critical infrastructures. Dan is the recipient of the NSF CAREER Award the IISE Transactions Best Application Paper Award. She has been recognized in multiple Best Track Paper and Best Student Paper Awards in Energy Systems, DAIS, and QCRE divisions at the IISE Annual Meetings.
Andi Wang
Andi Wang received his BS in statistics from Peking University in 2012 and his PhD in industrial engineering from Hong Kong University of Science and Technology in 2016. He also received his M.S. in computer science and engineering and another PhD in industrial engineering (system informatics and control) from Georgia Institute of Technology in 2021. He has been an assistant professor in the School of Manufacturing Systems and Networks in the Ira A. Fulton Schools of Engineering, Arizona State University from 2021-2024, and he joins the University of Wisconsin-Madison as an assistant professor in Fall 2024. Andi Wang’s research focuses on the intersection of data science and manufacturing systems. His research involves applying machine learning, high-dimensional statistics, and advanced optimization techniques to solve the challenges in manufacturing systems, and perform root-cause diagnostics, monitoring, design optimization, prediction for complex, interconnected, and intelligent systems. He is a recipient of a Wayne Kay Scholarship from SME, a recipient of INFORMS 2019 Data Mining Best Paper Finalist Award, INFORMS 2020 Quality Reliability and Statistics Best Paper Finalist Award, IISE QCRE 2021 Best Paper Finalist Award.