Matches in SemOpenAlex for { <https://semopenalex.org/work/W4300183416> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4300183416 abstract "Given a collection of data points, non-negative matrix factorization (NMF) suggests to express them as convex combinations of a small set of `archetypes' with non-negative entries. This decomposition is unique only if the true archetypes are non-negative and sufficiently sparse (or the weights are sufficiently sparse), a regime that is captured by the separability condition and its generalizations. In this paper, we study an approach to NMF that can be traced back to the work of Cutler and Breiman (1994) and does not require the data to be separable, while providing a generally unique decomposition. We optimize the trade-off between two objectives: we minimize the distance of the data points from the convex envelope of the archetypes (which can be interpreted as an empirical risk), while minimizing the distance of the archetypes from the convex envelope of the data (which can be interpreted as a data-dependent regularization). The archetypal analysis method of (Cutler, Breiman, 1994) is recovered as the limiting case in which the last term is given infinite weight. We introduce a `uniqueness condition' on the data which is necessary for exactly recovering the archetypes from noiseless data. We prove that, under uniqueness (plus additional regularity conditions on the geometry of the archetypes), our estimator is robust. While our approach requires solving a non-convex optimization problem, we find that standard optimization methods succeed in finding good solutions both for real and synthetic data." @default.
- W4300183416 created "2022-10-03" @default.
- W4300183416 creator A5011999109 @default.
- W4300183416 creator A5042744075 @default.
- W4300183416 date "2017-05-08" @default.
- W4300183416 modified "2023-09-28" @default.
- W4300183416 title "Non-negative Matrix Factorization via Archetypal Analysis" @default.
- W4300183416 doi "https://doi.org/10.48550/arxiv.1705.02994" @default.
- W4300183416 hasPublicationYear "2017" @default.
- W4300183416 type Work @default.
- W4300183416 citedByCount "0" @default.
- W4300183416 crossrefType "posted-content" @default.
- W4300183416 hasAuthorship W4300183416A5011999109 @default.
- W4300183416 hasAuthorship W4300183416A5042744075 @default.
- W4300183416 hasBestOaLocation W43001834161 @default.
- W4300183416 hasConcept C106159729 @default.
- W4300183416 hasConcept C106487976 @default.
- W4300183416 hasConcept C112680207 @default.
- W4300183416 hasConcept C11413529 @default.
- W4300183416 hasConcept C121332964 @default.
- W4300183416 hasConcept C126255220 @default.
- W4300183416 hasConcept C134306372 @default.
- W4300183416 hasConcept C152671427 @default.
- W4300183416 hasConcept C154945302 @default.
- W4300183416 hasConcept C157972887 @default.
- W4300183416 hasConcept C158693339 @default.
- W4300183416 hasConcept C159985019 @default.
- W4300183416 hasConcept C160920958 @default.
- W4300183416 hasConcept C162324750 @default.
- W4300183416 hasConcept C163716315 @default.
- W4300183416 hasConcept C192562407 @default.
- W4300183416 hasConcept C2524010 @default.
- W4300183416 hasConcept C2776135515 @default.
- W4300183416 hasConcept C2777021972 @default.
- W4300183416 hasConcept C33923547 @default.
- W4300183416 hasConcept C41008148 @default.
- W4300183416 hasConcept C42355184 @default.
- W4300183416 hasConcept C56372850 @default.
- W4300183416 hasConcept C62520636 @default.
- W4300183416 hasConcept C70710897 @default.
- W4300183416 hasConcept C72134830 @default.
- W4300183416 hasConceptScore W4300183416C106159729 @default.
- W4300183416 hasConceptScore W4300183416C106487976 @default.
- W4300183416 hasConceptScore W4300183416C112680207 @default.
- W4300183416 hasConceptScore W4300183416C11413529 @default.
- W4300183416 hasConceptScore W4300183416C121332964 @default.
- W4300183416 hasConceptScore W4300183416C126255220 @default.
- W4300183416 hasConceptScore W4300183416C134306372 @default.
- W4300183416 hasConceptScore W4300183416C152671427 @default.
- W4300183416 hasConceptScore W4300183416C154945302 @default.
- W4300183416 hasConceptScore W4300183416C157972887 @default.
- W4300183416 hasConceptScore W4300183416C158693339 @default.
- W4300183416 hasConceptScore W4300183416C159985019 @default.
- W4300183416 hasConceptScore W4300183416C160920958 @default.
- W4300183416 hasConceptScore W4300183416C162324750 @default.
- W4300183416 hasConceptScore W4300183416C163716315 @default.
- W4300183416 hasConceptScore W4300183416C192562407 @default.
- W4300183416 hasConceptScore W4300183416C2524010 @default.
- W4300183416 hasConceptScore W4300183416C2776135515 @default.
- W4300183416 hasConceptScore W4300183416C2777021972 @default.
- W4300183416 hasConceptScore W4300183416C33923547 @default.
- W4300183416 hasConceptScore W4300183416C41008148 @default.
- W4300183416 hasConceptScore W4300183416C42355184 @default.
- W4300183416 hasConceptScore W4300183416C56372850 @default.
- W4300183416 hasConceptScore W4300183416C62520636 @default.
- W4300183416 hasConceptScore W4300183416C70710897 @default.
- W4300183416 hasConceptScore W4300183416C72134830 @default.
- W4300183416 hasLocation W43001834161 @default.
- W4300183416 hasOpenAccess W4300183416 @default.
- W4300183416 hasPrimaryLocation W43001834161 @default.
- W4300183416 hasRelatedWork W1497354029 @default.
- W4300183416 hasRelatedWork W1528932392 @default.
- W4300183416 hasRelatedWork W1704327882 @default.
- W4300183416 hasRelatedWork W2098176827 @default.
- W4300183416 hasRelatedWork W2115221929 @default.
- W4300183416 hasRelatedWork W2162130060 @default.
- W4300183416 hasRelatedWork W2611997107 @default.
- W4300183416 hasRelatedWork W3115589710 @default.
- W4300183416 hasRelatedWork W3126264452 @default.
- W4300183416 hasRelatedWork W4300183416 @default.
- W4300183416 isParatext "false" @default.
- W4300183416 isRetracted "false" @default.
- W4300183416 workType "article" @default.