Embedding and projection

available

Construct lower-dimensional representations for visualization, comparison, and embedded-space clustering.

Overview

Embedding and projection methods transform selected dataset features into a lower-dimensional coordinate space.

The resulting representation can be inspected directly, compared across runs, used for two- or three-dimensional visualization, or supplied as the analytical space for clustering.

Supported methods

Different representation methods preserve or emphasize different aspects of dataset structure.

Linear projection

PCA

available

Constructs orthogonal components that capture decreasing amounts of dataset variance.

Linear Projection

Nonlinear embedding

UMAP

available

Constructs a lower-dimensional representation based on neighborhood relationships.

Nonlinear Neighborhood based

t-SNE

available

Creates a representation emphasizing local similarity between observations.

Nonlinear Local structure

MDS

available

Places observations in a lower-dimensional space while approximating pairwise distances.

Distance based Projection

Neural representation

Autoencoder

available

Learns a compressed representation through a neural encoder-decoder architecture.

Neural Learned representation

Variational Autoencoder

available

Learns a probabilistic latent representation using a variational encoder-decoder architecture.

Neural Probabilistic

DEC

available

Uses a learned latent representation designed around deep embedded structure.

Neural Embedded structure

How embeddings are used

Visualization

Inspect the relative arrangement of observations in two or three dimensions.

Embedded-space clustering

Use generated coordinates as the analytical input to a clustering method.

Learn more →

Cross-run comparison

Compare how alternative embedding methods and configurations represent the same dataset.

Persisted artifacts

Preserve generated coordinates together with their run configuration and downstream outputs.

Interpretation

Visual separation in an embedding is not automatically evidence of a meaningful real-world group.
Different methods preserve and emphasize different kinds of relationships.
Results can depend on feature selection, preprocessing, parameters, and random initialization.
Distances and apparent group sizes do not necessarily have equivalent meaning across embedding methods.