Provable ICA with Unknown Gaussian Noise, and Implications for Gaussian Mixtures and Autoencoders
Author(s): Arora, Sanjeev; Ge, Rong; Moitra, Ankur; Sachdeva, Sushant
DownloadTo refer to this page use:
http://arks.princeton.edu/ark:/88435/pr1885k
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Arora, Sanjeev | - |
dc.contributor.author | Ge, Rong | - |
dc.contributor.author | Moitra, Ankur | - |
dc.contributor.author | Sachdeva, Sushant | - |
dc.date.accessioned | 2021-10-08T19:49:52Z | - |
dc.date.available | 2021-10-08T19:49:52Z | - |
dc.date.issued | 2015 | en_US |
dc.identifier.citation | Arora, Sanjeev, Rong Ge, Ankur Moitra, and Sushant Sachdeva. "Provable ICA with Unknown Gaussian Noise, and Implications for Gaussian Mixtures and Autoencoders." Algorithmica 72, no. 1 (2015): 215-236. doi:10.1007/s00453-015-9972-2 | en_US |
dc.identifier.issn | 0178-4617 | - |
dc.identifier.uri | https://dspace.mit.edu/bitstream/handle/1721.1/106898/453_2015_9972_ReferencePDF.pdf?sequence=2&isAllowed=y | - |
dc.identifier.uri | http://arks.princeton.edu/ark:/88435/pr1885k | - |
dc.description.abstract | We present a new algorithm for independent component analysis which has provable performance guarantees. In particular, suppose we are given samples of the form π¦=π΄π₯+π where π΄ is an unknown but non-singular πΓπ matrix, π₯ is a random variable whose coordinates are independent and have a fourth order moment strictly less than that of a standard Gaussian random variable and π is an π-dimensional Gaussian random variable with unknown covariance π΄: We give an algorithm that provably recovers π΄ and π΄ up to an additive π and whose running time and sample complexity are polynomial in π and 1/π. To accomplish this, we introduce a novel βquasi-whiteningβ step that may be useful in other applications where there is additive Gaussian noise whose covariance is unknown. We also give a general framework for finding all local optima of a function (given an oracle for approximately finding just one) and this is a crucial step in our algorithm, one that has been overlooked in previous attempts, and allows us to control the accumulation of error when we find the columns of π΄ one by one via local search. | en_US |
dc.format.extent | 215 - 236 | en_US |
dc.language.iso | en_US | en_US |
dc.relation.ispartof | Algorithmica | en_US |
dc.rights | Author's manuscript | en_US |
dc.title | Provable ICA with Unknown Gaussian Noise, and Implications for Gaussian Mixtures and Autoencoders | en_US |
dc.type | Journal Article | en_US |
dc.identifier.doi | 10.1007/s00453-015-9972-2 | - |
dc.identifier.eissn | 1432-0541 | - |
pu.type.symplectic | http://www.symplectic.co.uk/publications/atom-terms/1.0/journal-article | en_US |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
ProvableICA.pdf | 417.26 kB | Adobe PDF | View/Download |
Items in OAR@Princeton are protected by copyright, with all rights reserved, unless otherwise indicated.