PFL-NON-IID Framework: Evaluating MOON Algorithm on Handling Non-IID Data Distributions

DOI

10.2991/978-94-6463-300-9_23How to use a DOI?

Keywords

Deep learning; Moon; Data models; Federated learning; Training effectiveness

Abstract

This paper introduces an optimized version of the MOON algorithm for federated learning, which is a distributed machine learning approach designed to tackle challenges related to data privacy and decentralization. However, the performance of traditional federated learning methods is hindered by the non-uniform distribution of data across nodes, a problem known as the Non-Identical and Independently Distributed (NON-IID) problem. The MOON algorithm leverages statistically heterogeneous data for personalized model training and improves the issues of slow gradient descent rates and high network communication overhead. Although the MOON algorithm has shown promising results, it still faces challenges in terms of computational complexity and training efficiency. Therefore, this paper aims to further optimize the MOON algorithm to enhance its computational efficiency and training effectiveness. The paper implements the MOON algorithm using the PFL-NON-IID framework and verifies its effectiveness in handling non-uniform data distributions. It also analyzes and compares the experimental results to optimize the algorithm's computational performance, real-time capability, and scalability. The study aims to provide support for the application of distributed deep learning systems.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Sheng Chen
AU  - Jiancheng Peng
AU  - Andi Tong
AU  - Cong Wu
PY  - 2023
DA  - 2023/11/27
TI  - PFL-NON-IID Framework: Evaluating MOON Algorithm on Handling Non-IID Data Distributions
BT  - Proceedings of the 2023 International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2023)
PB  - Atlantis Press
SP  - 224
EP  - 235
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-300-9_23
DO  - 10.2991/978-94-6463-300-9_23
ID  - Chen2023
ER  -