t-SNE Exaggerates Clusters, Provably
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912936250834944 |
|---|---|
| author | Bergam, Noah Snoeck, Szymon Verma, Nakul |
| author_facet | Bergam, Noah Snoeck, Szymon Verma, Nakul |
| contents | Central to the widespread use of t-distributed stochastic neighbor embedding (t-SNE) is the conviction that it produces visualizations whose structure roughly matches that of the input. To the contrary, we prove that (1) the strength of the input clustering, and (2) the extremity of outlier points, cannot be reliably inferred from the t-SNE output. We demonstrate the prevalence of these failure modes in practice as well. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_07746 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | t-SNE Exaggerates Clusters, Provably Bergam, Noah Snoeck, Szymon Verma, Nakul Machine Learning Central to the widespread use of t-distributed stochastic neighbor embedding (t-SNE) is the conviction that it produces visualizations whose structure roughly matches that of the input. To the contrary, we prove that (1) the strength of the input clustering, and (2) the extremity of outlier points, cannot be reliably inferred from the t-SNE output. We demonstrate the prevalence of these failure modes in practice as well. |
| title | t-SNE Exaggerates Clusters, Provably |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2510.07746 |