Shiqing-Liu/Privacy-Preserving-Federated-Learning

392274 ISY Project: Privacy-Preserving Federated Learning (Pj) (SoSe 2022)

Privacy-Preserving Federated Learning

This repository contains code for the ISY project 'Privacy-Preserving Federated Learning', which won the Best Project Award🥇 at the 2022 Congress of the Master's degree program in Intelligent Systems.

Team members: Hela Jemli, Fabian Luca Reichwald, and Louis Scheu

Supervised by: Prof. Yaochu Jin, Yuping Yan, and Shiqing Liu

More info: https://blogs.uni-bielefeld.de/blog/techfak/entry/isy_kongress_k%C3%BCrt_bestes_studi1

Overview

Our project "Privacy-Preserving Federated Learning" received the "Best Project Award" during the Congress of Intelligent Systems in November 2022 by an expert jury. The goal was to investigate the uprising technology of Federated Learning (in which Machine Learning is performed decentralized on edge devices such as mobile phones in connection with a central server) concerning data privacy, computational, and communication efficiency. After a systematical investigation, we combined several high-performing solutions to each problem and created a modular system to mimic real-world scenarios of unstable and unbalanced connectivity. The involved team members were Hela Jemli, Fabian Luca Reichwald, and Louis Scheu. The project was supervised by Prof. Yaochu Jin, Yuping Yan, and Shiqing Liu.

Related Sources

Survey papers

• Federated Learning: Challenges, methods, and future directions

• Federated Machine Learning: Concept and Applications

Baseline algorithms

• FedAvg Communication-Efficient Learning of Deep Networks from Decentralized Data

• FedOpt Adaptive Federated Optimization

Communication efficiency

• Ternary Compression for Communication-Efficient Federated Learning

• Communication-efficient federated learning.

• HCFL: A High Compression Approach for Communication-Efficient Federated Learning in Very Large Scale IoT Networks

Non-IID distribution

• Federated Learning on Non-IID Data: A Survey

• Federated Learning with Personalization Layers

• Personalized Federated Learning Code on Flickr and Cifar

Useful links

• https://github.com/chaoyanghe/Awesome-Federated-Learning

• https://ai.googleblog.com/2017/04/federated-learning-collaborative.html

• https://towardsdatascience.com/federated-learning-your-data-stays-with-you-fc1d49b35ec4

• https://towardsdatascience.com/preserving-data-privacy-in-deep-learning-part-1-a04894f78029