SQuadMDS: A lean Stochastic Quartet MDS improving global structure preservation in neighbor embedding like t-SNE and UMAP

Multidimensional scaling is a process that aims to embed high dimensional data into a lower-dimensional space; this process is often used for the purpose of data visualisation. Common multidimensional scaling algorithms tend to have high computational complexities, making them inapplicable on large...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Neurocomputing (Amsterdam) Ročník 503; s. 17 - 27
Hlavní autoři: Lambert, Pierre, de Bodt, Cyril, Verleysen, Michel, Lee, John A.
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 07.09.2022
Témata:
Data visualization
Dimensionality reduction
Distance preservation
Global structure
MDS
Neighbor embedding
Stochastic algorithms
t-SNE
UMAP
Dimensionality reduction
UMAP
Stochastic algorithms
Neighbor embedding
t-SNE
DR
LD/HD space
SGD
Data visualization
Global structure
MDS
Distance preservation
mMDS
ISSN:0925-2312, 1872-8286
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Multidimensional scaling is a process that aims to embed high dimensional data into a lower-dimensional space; this process is often used for the purpose of data visualisation. Common multidimensional scaling algorithms tend to have high computational complexities, making them inapplicable on large data sets. This work introduces a stochastic, force directed approach to multidimensional scaling with a time and space complexity of O(N), with N data points. The method can be combined with force directed layouts of the family of neighbour embedding such as t-SNE, to produce embeddings that preserve both the global and the local structures of the data. Experiments assess the quality of the embeddings produced by the standalone version and its hybrid extension both quantitatively and qualitatively, showing competitive results outperforming state-of-the-art approaches. Codes are available athttps://github.com/PierreLambert3/SQuaD-MDS-and-FItSNE-hybrid.
ISSN:0925-2312
1872-8286
DOI:10.1016/j.neucom.2022.06.108