Mostra el registre d'ítem simple
A new spectral method for latent variable models
dc.contributor.author | Ruffini, Matteo |
dc.contributor.author | Casanellas Rius, Marta |
dc.contributor.author | Gavaldà Mestre, Ricard |
dc.contributor.other | Universitat Politècnica de Catalunya. Departament de Matemàtiques |
dc.contributor.other | Universitat Politècnica de Catalunya. Departament de Ciències de la Computació |
dc.date.accessioned | 2018-10-03T09:38:18Z |
dc.date.available | 2019-09-01T00:25:44Z |
dc.date.issued | 2018-05-22 |
dc.identifier.citation | Ruffini, M., Casanellas, M., Gavaldà, R. A new spectral method for latent variable models. "Machine learning", 22 Maig 2018, vol. 107, núm. 8-10, p. 1431-1455. |
dc.identifier.issn | 0885-6125 |
dc.identifier.other | https://arxiv.org/abs/1612.03409 |
dc.identifier.uri | http://hdl.handle.net/2117/121828 |
dc.description.abstract | We present an algorithm for the unsupervised learning of latent variable models based on the method of moments. We give efficient estimates of the moments for two models that are well known, e.g., in text mining, the single-topic model and latent Dirichlet allocation, and we provide a tensor decomposition algorithm for the moments that proves to be robust both in theory and in practice. Experiments on synthetic data show that the proposed estimators outperform the existing ones in terms of reconstruction accuracy, and that the proposed tensor decomposition technique achieves the learning accuracy of the state-of-the-art method with significantly smaller running times. We also provide examples of applications to real-world text corpora for both single-topic model and LDA, obtaining meaningful results |
dc.format.extent | 25 p. |
dc.language.iso | eng |
dc.rights | Attribution-NonCommercial-NoDerivs 3.0 Spain |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/3.0/es/ |
dc.subject | Àrees temàtiques de la UPC::Matemàtiques i estadística |
dc.subject.lcsh | Mathematical statistics |
dc.subject.lcsh | Numerical analysis |
dc.subject.other | Spectral methods |
dc.subject.other | Method of moments |
dc.subject.other | Latent variable models |
dc.subject.other | Topic modeling |
dc.title | A new spectral method for latent variable models |
dc.type | Article |
dc.subject.lemac | Estadística matemàtica |
dc.subject.lemac | Anàlisi numèrica |
dc.subject.lemac | Anàlisi espectral |
dc.contributor.group | Universitat Politècnica de Catalunya. GEOMVAP - Geometria de Varietats i Aplicacions |
dc.contributor.group | Universitat Politècnica de Catalunya. LARCA - Laboratori d'Algorísmia Relacional, Complexitat i Aprenentatge |
dc.identifier.doi | 10.1007/s10994-018-5706-4 |
dc.description.peerreviewed | Peer Reviewed |
dc.relation.publisherversion | https://link.springer.com/article/10.1007%2Fs10994-018-5706-4 |
dc.rights.access | Open Access |
local.identifier.drac | 23356244 |
dc.description.version | Postprint (author's final draft) |
local.citation.author | Ruffini, M.; Casanellas, M.; Gavaldà, R. |
local.citation.publicationName | Machine learning |
local.citation.volume | 107 |
local.citation.number | 8-10 |
local.citation.startingPage | 1431 |
local.citation.endingPage | 1455 |
Fitxers d'aquest items
Aquest ítem apareix a les col·leccions següents
-
Articles de revista [134]
-
Articles de revista [184]
-
Articles de revista [1.049]
-
Articles de revista [3.268]