UMAP: Uniform Manifold Approximation and Projection

@article{McInnes2018UMAPUM,
  title={UMAP: Uniform Manifold Approximation and Projection},
  author={L. McInnes and John Healy and Nathaniel Saul and Lukas Gro{\ss}berger},
  journal={J. Open Source Softw.},
  year={2018},
  volume={3},
  pages={861}
}
Uniform Manifold Approximation and Projection (UMAP) is a dimension reduction technique that can be used for visualisation similarly to t-SNE, but also for general non-linear dimension reduction. UMAP has a rigorous mathematical foundation, but is simple to use, with a scikit-learn compatible API. UMAP is among the fastest manifold learning implementations available – significantly faster than most t-SNE implementations. 
The mathematics of UMAP
Unsupervised Functional Data Analysis via Nonlinear Dimension Reduction
Normalizing Flows Across Dimensions
Quadric hypersurface intersection for manifold learning in feature space
Multi-objective genetic programming for manifold learning: balancing quality and dimensionality
Neighborhood Normalization for Robust Geometric Feature Learning – Supplementary Material
Unsupervised Sentence-embeddings by Manifold Approximation and Projection
...
1
2
3
4
5
...

References