1 / 17

Non-Linear Dimensionality Reduction

Non-Linear Dimensionality Reduction. Presented by: Johnathan Franck Mentor: Alex Cloninger. Outline. Different Representations 5 Techniques Principal component analysis (PCA)/ Multi-dimensional scaling (MDS) Sammons non-linear mapping Isomap Kernel PCA Locally linear embedding.

cleave
Download Presentation

Non-Linear Dimensionality Reduction

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Non-Linear Dimensionality Reduction Presented by: Johnathan Franck Mentor: Alex Cloninger

  2. Outline • Different Representations • 5 Techniques • Principal component analysis (PCA)/Multi-dimensional scaling (MDS) • Sammons non-linear mapping • Isomap • Kernel PCA • Locally linear embedding

  3. Different viewpoints

  4. Principal Component Analysis/Multidimensional Scaling Singular Value Decomposition of A:

  5. PCA/MDS Formulation X – reduced dimension Y – higher dimension W - Orthogonal Singular Value decomposition of Y equivalent to Eigenvalue decomposition of gram matrix Preserve Gram Matrix: S Optimization: Linear Algebra Distance Preserved Rotation

  6. Swiss Roll Open Box

  7. Sammons Nonlinear mapping Weight distances differently

  8. Swiss Roll Open Box

  9. Isomap Approximate geodesic • Build graph with K-neighbors/ε-ball • Weight graph with Euclidean distance • Compute pairwise distances with Dijkstra’s algorithm

  10. Swiss Roll Open Box

  11. Kernel PCA What other distance functions? Mercer’s theorem

  12. Swiss Roll Open Box

  13. Local Linear Embedding Preserve relation to local points

  14. Swiss Roll Open Box

  15. Isomap Original 3 dimensions PCA Kernel PCA Sammon’s nonlinear mapping Locally Linear Embedding

  16. Thanks • Mentor: Alex Cloninger • DRP program • Contact information: • Jafranck2@gmail.com

  17. Sources • Images • Nonlinear Dimensionality reduction • By John Lee and Michael Verleyson • http://blog.pluskid.org/wp-content/uploads/2010/05/isomap-graph.png • http://imagineatness.files.wordpress.com/2011/12/gaussian.png • http://scikit-learn.org/0.11/_images/plot_kernel_pca_1.png • http://www.cs.nyu.edu/~roweis/lle/images/llef2med.gif • http://upload.wikimedia.org/wikipedia/commons/b/bb/SOMsPCA.PNG

More Related