Dimensionality reduction in decentralized networks by Gossip aggregation of principal components analyzers.
Jérôme FellusDavid PicardPhilippe Henri GosselinPublished in: ESANN (2014)
Keyphrases
- principal components
- dimensionality reduction
- principal component analysis
- structured peer to peer
- multivariate data
- overlay network
- high dimensional data
- high dimensional
- low dimensional
- principal components analysis
- hyperplane
- feature extraction
- social networks
- feature space
- network structure
- kernel space
- multi agent
- data points
- manifold learning
- pattern recognition
- distributed systems
- peer to peer
- metric learning
- feature set
- principal component regression
- linear discriminant analysis
- singular value decomposition
- data sets
- input space
- computer vision
- dimensionality reduction methods
- nearest neighbor
- machine learning