Client Clustering

Client clustering in federated learning aims to improve the efficiency and accuracy of collaborative model training across decentralized devices with heterogeneous data distributions. Current research focuses on developing algorithms that effectively group clients based on data similarity, often leveraging model weights or prediction uncertainty, without requiring extensive communication or pre-defined cluster numbers. These advancements address the significant challenge of non-independent and identically distributed (non-IID) data, leading to improved model performance and reduced training time in various applications, including fault diagnosis and multimedia processing.

Papers

October 20, 2024

Bayesian data fusion for distributed learning
Peng Wu, Tales Imbiriba, Pau Closas
LeArning Abstract Cluster Structure Probabilistic Fusion Data Association Client Clustering

July 9, 2024

FedClust: Tackling Data Heterogeneity in Federated Learning through Weight-Driven Client Clustering
Md Sirajul Islam, Simin Javaherian, Fei Xu, Xu Yuan, Li Chen, Nian-Feng Tzeng
Data Heterogeneity Model Weight Machine Learning Paradigm Client Clustering

June 18, 2024

SFedCA: Credit Assignment-Based Active Client Selection Strategy for Spiking Federated Learning
Qiugang Zhan, Jinbo Cao, Xiurui Xie, Malu Zhang, Huajin Tang, Guisong Liu
Federated Learning Client Selection Data Distribution Client Clustering Consumer Credit Global Sampling

June 12, 2024

Regularizing and Aggregating Clients with Class Distribution for Personalized Federated Learning
Gyuejeong Lee, Daeyoung Choi
Personalized Federated Learning Federated Averaging Neural Network Weight Class Distribution Client Clustering

March 7, 2024

FedClust: Optimizing Federated Learning on Non-IID Data through Weight-Driven Client Clustering
Md Sirajul Islam, Simin Javaherian, Fei Xu, Xu Yuan, Li Chen, Nian-Feng Tzeng
Model Training Client Clustering

November 28, 2023

Contrastive encoder pre-training-based clustered federated learning for heterogeneous data
Ye Lin Tun, Minh N. H. Nguyen, Chu Myaet Thwal, Jinwoo Choi, Choong Seon Hong
LeArning Abstract Contrastive Learning Heterogeneous Data Client Clustering Contrastive Encoder

September 25, 2023

REPA: Client Clustering without Training and Data Labels for Improved Federated Learning in Non-IID Settings
Boris Radovič, Veljko Pejović
Training Data Federated Learning Autoencoder Based Data Labeling Client Clustering

August 30, 2023

Federated Two Stage Decoupling With Adaptive Personalization Layers
Hangyu Zhu, Yuxiang Fan, Zhenping Xie
Data Heterogeneity Federated Prompt Cooperation Cluster Structure Client Heterogeneity Light Stage Client Clustering Personalization Layer

April 26, 2023

Federated Learning with Uncertainty-Based Client Clustering for Fleet-Wide Fault Diagnosis
Hao Lu, Adam Thelen, Olga Fink, Chao Hu, Simon Laflamme
Fault Diagnosis Decentralized Agent Industrial Monitoring Client Clustering Diagnosis Algorithm

March 2, 2023

Stochastic Clustered Federated Learning
Dun Zeng, Xiangjing Hu, Shiyu Liu, Yue Yu, Qifan Wang, Zenglin Xu
Learning Framework Efficient Baseline Client Clustering

October 29, 2022

Auxo: Efficient Federated Learning via Scalable Client Clustering
Jiachen Liu, Fan Lai, Yinwei Dai, Aditya Akella, Harsha Madhyastha, Mosharaf Chowdhury
Federated Learning Client Heterogeneity Client Clustering Cohort Representation Cohort Comfort Model

October 11, 2022

Client Error Clustering Approaches in Content Delivery Networks (CDN)
Ermiyas Birihanu, Jiyan Mahmud, Péter Kiss, Adolf Kamuzora, Wadie Skaf, Tomáš Horváth, Tamás Jursonovics, Peter Pogrzeba, Imre Lendák
Client Clustering

June 17, 2022

Decentralized adaptive clustering of deep nets is beneficial for client collaboration
Edvin Listo Zec, Ebba Ekblom, Martin Willbo, Olof Mogren, Sarunas Girdzijauskas
Deep Network Decentralized FL Gossip Algorithm Local Learning Client Collaboration Peer to Peer Adaptive Clustering Soft Clustering Client Clustering

Client Clustering

Papers

Bayesian data fusion for distributed learning

FedClust: Tackling Data Heterogeneity in Federated Learning through Weight-Driven Client Clustering

SFedCA: Credit Assignment-Based Active Client Selection Strategy for Spiking Federated Learning

Regularizing and Aggregating Clients with Class Distribution for Personalized Federated Learning

FedClust: Optimizing Federated Learning on Non-IID Data through Weight-Driven Client Clustering

Contrastive encoder pre-training-based clustered federated learning for heterogeneous data

REPA: Client Clustering without Training and Data Labels for Improved Federated Learning in Non-IID Settings

Federated Two Stage Decoupling With Adaptive Personalization Layers

Federated Learning with Uncertainty-Based Client Clustering for Fleet-Wide Fault Diagnosis

Stochastic Clustered Federated Learning

Auxo: Efficient Federated Learning via Scalable Client Clustering

Client Error Clustering Approaches in Content Delivery Networks (CDN)

Decentralized adaptive clustering of deep nets is beneficial for client collaboration