This paper shows that regular, unsupervised t-SNE can be turned into a class- aware embedding method, coined as cat-SNE. Class labels are accounted in the.

The python code catsne.py implements cat-SNE, a class-aware version of t-SNE, as well as quality assessment criteria for both supervised and unsupervised ...

Experimental results show that the proposed class-aware t -SNE (ca t -SNE) outperforms regular t -SNE in K NN classification tasks carried out in the ...

This paper proposes a modification of t-SNE that employs class labels to adjust the widths of the Gaussian neighborhoods around each datum, instead of deriving ...

Jan 3, 2021 · Experimental results show that the proposed class-aware t-SNE (cat-SNE) outperforms regular t-SNE in KNN classification tasks carried out in the ...

The blue social bookmark and publication sharing system.

Apr 7, 2017 · The result of t-SNE separates classes very well. This implies that it is possible to build classification model that will also separate classes very well.

Missing: cat- | Show results with:cat-

Class-aware t-SNE: Cat-SNE. de Bodt, C. Mulders, D. López-Sánchez, D. Verleysen, M. Lee, J.A.. Proceedings: ESANN 2019 - Proceedings, 27th European Symposium ...

Sep 14, 2023 · t-SNE preservation of global and local neighborhoods for unlabeled data and the cat-SNE capability to cluster instances sharing the same label.