Towards One Model for Classical Dimensionality Reduction: A Probabilistic Perspective on UMAP and t-SNE

Aditya Ravuri; Neil D. Lawrence

Back to publications

Towards One Model for Classical Dimensionality Reduction: A Probabilistic Perspective on UMAP and t-SNE

Aditya Ravuri, Neil D. Lawrence

, 2024.

Abstract

This paper shows that the dimensionality reduction methods, UMAP and t-SNE, can be approximately recast as MAP inference methods corresponding to a generalized Wishart-based model introduced in ProbDR. This interpretation offers deeper theoretical insights into these algorithms, while introducing tools with which similar dimensionality reduction methods can be studied.

Links

Cite this Paper

BibTeX


@Misc{publications/towards-one-model-for-classical-dimensionality-reduction,
  title = 	 {Towards One Model for Classical Dimensionality Reduction: A Probabilistic Perspective on UMAP and t-SNE},
  author = 	 {Ravuri, Aditya and Lawrence, Neil D.},
  year = 	 {2024},
  url = 	 {/publications/towards-one-model-for-classical-dimensionality-reduction.html},
  abstract = 	 {This paper shows that the dimensionality reduction methods, UMAP and t-SNE, can be approximately recast as MAP inference methods corresponding to a generalized Wishart-based model introduced in ProbDR. This interpretation offers deeper theoretical insights into these algorithms, while introducing tools with which similar dimensionality reduction methods can be studied.
}
}

Endnote

%0 Generic
%T Towards One Model for Classical Dimensionality Reduction: A Probabilistic Perspective on UMAP and t-SNE
%A Aditya Ravuri
%A Neil D. Lawrence
%D 2024	
%F publications/towards-one-model-for-classical-dimensionality-reduction
%U /publications/towards-one-model-for-classical-dimensionality-reduction.html
%X This paper shows that the dimensionality reduction methods, UMAP and t-SNE, can be approximately recast as MAP inference methods corresponding to a generalized Wishart-based model introduced in ProbDR. This interpretation offers deeper theoretical insights into these algorithms, while introducing tools with which similar dimensionality reduction methods can be studied.

RIS


TY  - GEN
TI  - Towards One Model for Classical Dimensionality Reduction: A Probabilistic Perspective on UMAP and t-SNE
AU  - Aditya Ravuri
AU  - Neil D. Lawrence
DA  - 2024/05/27	
ID  - publications/towards-one-model-for-classical-dimensionality-reduction
UR  - /publications/towards-one-model-for-classical-dimensionality-reduction.html
AB  - This paper shows that the dimensionality reduction methods, UMAP and t-SNE, can be approximately recast as MAP inference methods corresponding to a generalized Wishart-based model introduced in ProbDR. This interpretation offers deeper theoretical insights into these algorithms, while introducing tools with which similar dimensionality reduction methods can be studied.

ER  -

APA


Ravuri, A. & Lawrence, N.D.. (2024). Towards One Model for Classical Dimensionality Reduction: A Probabilistic Perspective on UMAP and t-SNE.  Available from /publications/towards-one-model-for-classical-dimensionality-reduction.html.