Dimensionality and Curvature Selection of Graph Embedding using Decomposed Normalized Maximum Likelihood Code-Length.
Ryo YukiAtsushi SuzukiKenji YamanishiPublished in: ICDM (2023)
Keyphrases
- normalized maximum likelihood
- code length
- graph embedding
- dimensionality reduction
- gray code
- change point detection
- spatially adaptive
- low dimensional
- high dimensional
- semi supervised
- discriminant analysis
- minimum description length
- multiscale
- feature space
- data points
- data sets
- high dimensional data
- mdl principle
- active learning
- feature selection
- channel coding
- outlier detection
- principal component analysis