Please use this identifier to cite or link to this item:
Title: Personalised federated learning with differential privacy and gradient selection
Authors: Lee, Jason Zhi Xin
Ng, Kai Chin
Toh, Arnold Xuan Ming
Keywords: Engineering::Computer science and engineering::Data
Issue Date: 2021
Publisher: Nanyang Technological University
Source: Lee, J. Z. X., Ng, K. C. & Toh, A. X. M. (2021). Personalised federated learning with differential privacy and gradient selection. Final Year Project (FYP), Nanyang Technological University, Singapore.
Abstract: The fast-emerging field of federated learning holds the promise of allowing clients to contribute to a central machine learning model without the need to send their data to a central server, thus providing privacy for their data. Two issues arise: dealing with statistical heterogeneity in datasets, which is often the case in real-world settings, and obtaining stricter data privacy through employing privacy-preserving mechanisms. In this paper, we use personalized layers in a federated Convolutional Neural Network model to address statistical heterogeneity and use differential privacy to provide mathematically rigorous privacy guarantees for the federated learning model. We also propose a gradient selection technique to increase model performance. We developed a framework combining these techniques and experimentally demonstrate the effectiveness of the proposed framework on a dataset in improving model performance whilst maintaining a reasonable level of privacy guarantee and training efficiency.
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:Renaissance Capstone Project (RCP)

Files in This Item:
File Description SizeFormat 
Final Report.pdf
  Restricted Access
1.1 MBAdobe PDFView/Open

Page view(s)

Updated on Jun 22, 2022


Updated on Jun 22, 2022

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.