학술논문

Privacy Leakage in Federated Home Applications Using Gradient Inversion Algorithms
Document Type
Conference
Source
2024 IEEE International Conference on Industrial Technology (ICIT) Industrial Technology (ICIT), 2024 IEEE International Conference on. :1-6 Mar, 2024
Subject
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Training
Knowledge engineering
Privacy
Data privacy
Voltage measurement
Machine learning algorithms
Federated learning
Privacy leakage
Gradient inversion
Home appliances
Language
ISSN
2643-2978
Abstract
With advances in smart metering infrastructure, household electricity metering data are remotely collected, leading to concerns about household privacy leakage. Federated learning is a promising solution because it avoids direct data uploading. However, recent research shows that the gradient of federated learning contains a certain amount of private information that can be recovered using gradient inversion algorithms. This paper proposes an explainable algorithm to locate the key factors affecting privacy leakage and vulnerable data in home application scenarios. Simulations show comparative results of privacy leakages in different situations and reveal that for home Artificial Intelligence applications, smaller batch sizes, training iterations, and extreme values are prone to causing privacy leaks. Based on that, the advice for protecting federated learning privacy under gradient inversion algorithms is summarized.