REPA: Client Clustering without Training and Data Labels for Improved Federated Learning in Non-IID Settings.
Boris RadovicVeljko PejovicPublished in: CoRR (2023)
Keyphrases
- data points
- online learning
- training data
- data sets
- original data
- training examples
- data sources
- data collection
- prior knowledge
- data analysis
- multidimensional data
- distributed data
- supervised learning
- data structure
- data objects
- learning algorithm
- training samples
- reinforcement learning
- central server
- categorical data
- cluster analysis
- high dimensional data
- learning process
- unsupervised learning
- database
- image data
- labeled data
- spectral clustering
- small number
- positive examples
- digital libraries
- clustering analysis
- probability distribution
- labelled data