Manifold learning
Manifold learning is a family of dimensionality-reduction methods that models high-dimensional data as lying on or near a lower-dimensional manifold.
Build AI Drug Discovery Pipelines
Manifold learning is a family of dimensionality-reduction methods that models high-dimensional data as lying on or near a lower-dimensional manifold.
- How it is used
- Applies nonlinear dimensionality reduction to life-sciences datasets and compound-screening workflows.
- How deep it goes
- Explains the manifold assumption, contrasts PCA with nonlinear structure, and demonstrates tuned UMAP with classification.
- Examples
- Uses the Swiss roll, COVID Moonshot compounds, fingerprints, activity labels, and chemical-series projections.
- What you can do after
- Apply UMAP through a scikit-learn pipeline and evaluate parameters with cross-validation.
Read how Build AI Drug Discovery Pipelines explains manifold learning →
Retrieval Augmented Generation, The Foundational Ideas
Manifold learning is a dimensionality-reduction approach that models high-dimensional observations as lying near a lower-dimensional manifold.
- How it is used
- Supports clustering of embeddings within hierarchical retrieval pipelines.
- How deep it goes
- Defines neighborhood-preserving compression but does not detail the underlying mathematics or implementation steps.
- Examples
- Uses UMAP to compress high-dimensional embeddings into a few dimensions before clustering.
- What you can do after
- Apply it via UMAP to prepare embeddings for clustering in a retrieval system.
Read how Retrieval Augmented Generation, The Foundational Ideas explains manifold learning →
Related concepts
- UMAPincludes · UMAP, or Uniform Manifold Approximation and Projection, is a nonlinear manifold-learning technique that embeds high-dimensional data in a lower-dimensional space while preserving important neighborhood and structural relationships.
- principal component analysiscontrasts with · Principal component analysis is a linear statistical technique that represents data along orthogonal directions, called principal components, ordered by the variance they capture.
Questions about manifold learning
What is manifold learning?
Manifold learning is a family of dimensionality-reduction methods that models high-dimensional data as lying on or near a lower-dimensional manifold.
Which Manning books explain manifold learning?
Build AI Drug Discovery Pipelines and Retrieval Augmented Generation, The Foundational Ideas.
What is the difference between manifold learning and principal component analysis?
Manifold learning is a family of dimensionality-reduction methods that models high-dimensional data as lying on or near a lower-dimensional manifold. Principal component analysis is a linear statistical technique that represents data along orthogonal directions, called principal components, ordered by the variance they capture.
What kinds of manifold learning are there?
Kinds of manifold learning include UMAP.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

