Manifold-Aware Deep Clustering: Maximizing Angles Between Embedding Vectors Based on Regular Simplex.
Keitaro TanakaRyosuke SawataShusuke TakahashiPublished in: Interspeech (2021)
Keyphrases
- vector space
- clustering algorithm
- nonlinear dimensionality reduction
- manifold embedding
- clustering method
- binary vectors
- lower dimensional
- k means
- graph embedding
- low dimensional
- data points
- self organizing maps
- high dimensional
- feature space
- data clustering
- cluster analysis
- laplacian eigenmaps
- categorical data
- high dimensional data
- unsupervised learning
- linear programming
- locality preserving
- knn
- output space
- geodesic distance
- fuzzy clustering
- data sets
- locality preserving projections
- manifold learning
- spectral clustering
- information theoretic
- dimensionality reduction