Matches in SemOpenAlex for { <https://semopenalex.org/work/W2105724942> ?p ?o ?g. }
- W2105724942 endingPage "2832" @default.
- W2105724942 startingPage "2773" @default.
- W2105724942 abstract "This work considers a computationally and statistically efficient parameter estimation method for a wide class of latent variable models--including Gaussian mixture models, hidden Markov models, and latent Dirichlet allocation--which exploits a certain tensor structure in their low-order observable moments (typically, of second- and third-order). Specifically, parameter estimation is reduced to the problem of extracting a certain (orthogonal) decomposition of a symmetric tensor derived from the moments; this decomposition can be viewed as a natural generalization of the singular value decomposition for matrices. Although tensor decompositions are generally intractable to compute, the decomposition of these specially structured tensors can be efficiently obtained by a variety of approaches, including power iterations and maximization approaches (similar to the case of matrices). A detailed analysis of a robust tensor power method is provided, establishing an analogue of Wedin's perturbation theorem for the singular vectors of matrices. This implies a robust and computationally tractable estimation approach for several popular latent variable models." @default.
- W2105724942 created "2016-06-24" @default.
- W2105724942 creator A5018792915 @default.
- W2105724942 creator A5023215207 @default.
- W2105724942 creator A5035001911 @default.
- W2105724942 creator A5061246300 @default.
- W2105724942 creator A5075366851 @default.
- W2105724942 date "2014-01-01" @default.
- W2105724942 modified "2023-10-06" @default.
- W2105724942 title "Tensor decompositions for learning latent variable models" @default.
- W2105724942 cites W1535258871 @default.
- W2105724942 cites W1574816920 @default.
- W2105724942 cites W1584154100 @default.
- W2105724942 cites W1700863064 @default.
- W2105724942 cites W1755563775 @default.
- W2105724942 cites W1825640910 @default.
- W2105724942 cites W1827128995 @default.
- W2105724942 cites W18426229 @default.
- W2105724942 cites W1956647075 @default.
- W2105724942 cites W1969015668 @default.
- W2105724942 cites W1969082792 @default.
- W2105724942 cites W1980018091 @default.
- W2105724942 cites W1981367467 @default.
- W2105724942 cites W1982562599 @default.
- W2105724942 cites W1984151763 @default.
- W2105724942 cites W1992226551 @default.
- W2105724942 cites W1993529825 @default.
- W2105724942 cites W1995963238 @default.
- W2105724942 cites W2002598080 @default.
- W2105724942 cites W2010072035 @default.
- W2105724942 cites W2012828271 @default.
- W2105724942 cites W2014562510 @default.
- W2105724942 cites W2014565165 @default.
- W2105724942 cites W2018282388 @default.
- W2105724942 cites W2022852240 @default.
- W2105724942 cites W2024165284 @default.
- W2105724942 cites W2026302946 @default.
- W2105724942 cites W2033154334 @default.
- W2105724942 cites W2039748980 @default.
- W2105724942 cites W2046164006 @default.
- W2105724942 cites W2049633694 @default.
- W2105724942 cites W2049990581 @default.
- W2105724942 cites W2050616618 @default.
- W2105724942 cites W2057214765 @default.
- W2105724942 cites W2057503509 @default.
- W2105724942 cites W2057853309 @default.
- W2105724942 cites W2077223725 @default.
- W2105724942 cites W2079234446 @default.
- W2105724942 cites W2079705627 @default.
- W2105724942 cites W2090208105 @default.
- W2105724942 cites W2090799283 @default.
- W2105724942 cites W2095887647 @default.
- W2105724942 cites W2096295738 @default.
- W2105724942 cites W2097150846 @default.
- W2105724942 cites W2099565408 @default.
- W2105724942 cites W2099741732 @default.
- W2105724942 cites W2103581346 @default.
- W2105724942 cites W2106100072 @default.
- W2105724942 cites W2108911240 @default.
- W2105724942 cites W2113722075 @default.
- W2105724942 cites W2116658704 @default.
- W2105724942 cites W2117756735 @default.
- W2105724942 cites W2121142987 @default.
- W2105724942 cites W2121739212 @default.
- W2105724942 cites W2123649031 @default.
- W2105724942 cites W2124757684 @default.
- W2105724942 cites W2127218421 @default.
- W2105724942 cites W2128521126 @default.
- W2105724942 cites W2135342401 @default.
- W2105724942 cites W2139158168 @default.
- W2105724942 cites W2140756703 @default.
- W2105724942 cites W2141224535 @default.
- W2105724942 cites W2141986476 @default.
- W2105724942 cites W2142496058 @default.
- W2105724942 cites W2142975003 @default.
- W2105724942 cites W2146756121 @default.
- W2105724942 cites W2146758737 @default.
- W2105724942 cites W2149960632 @default.
- W2105724942 cites W2158055608 @default.
- W2105724942 cites W2158282517 @default.
- W2105724942 cites W2163246442 @default.
- W2105724942 cites W2208906386 @default.
- W2105724942 cites W2218986518 @default.
- W2105724942 cites W2257859587 @default.
- W2105724942 cites W2258054274 @default.
- W2105724942 cites W2292293189 @default.
- W2105724942 cites W2294384375 @default.
- W2105724942 cites W2330159389 @default.
- W2105724942 cites W2569661359 @default.
- W2105724942 cites W2798909945 @default.
- W2105724942 cites W2951177125 @default.
- W2105724942 cites W2952609396 @default.
- W2105724942 cites W2952995770 @default.
- W2105724942 cites W2953204889 @default.
- W2105724942 cites W2953337630 @default.
- W2105724942 cites W2962790067 @default.
- W2105724942 cites W2963491532 @default.
- W2105724942 cites W2963800155 @default.