Garduño Alvarado, Tzolkin; Sagols Troncoso, Feliú; Wolf, Gunnar; Soulier, Eddie y Rousseaux, Francis
(2025):
Dimension reduction algorithms and techniques.
Morfismos, 27 (1).
pp. 1-30.
ISSN 1870-6525
Resumen
The recent proliferation of multidimensional output models in AI has catalyzed advancements in dimension reduction research. While these models accommodate diverse application fields and dimensions, the sheer multiplicity doesn't always yield substantial insights. Dimension reduction emerges as a pivotal process for translating high-dimensional data into a more manageable lower-dimensional space. Achieving this requires preserving geometric and topological properties across both spaces. This article aims to explore fundamental dimension reduction techniques, essential for data science researchers venturing into this domain. Of particular focus is UMAP, currently recognized as the pinnacle of this field, offering an expansive research landscape for topologists, geometers, and mathematicians alike.
Acciones (requiere ser usuario registrado)
 |
Editar objeto |