Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387560894> ?p ?o ?g. }
Showing items 1 to 57 of
57
with 100 items per page.
- W4387560894 abstract "Streaming principal component analysis (PCA) is an integral tool in large-scale machine learning for rapidly estimating low-dimensional subspaces of very high dimensional and high arrival-rate data with missing entries and corrupting noise. However, modern trends increasingly combine data from a variety of sources, meaning they may exhibit heterogeneous quality across samples. Since standard streaming PCA algorithms do not account for non-uniform noise, their subspace estimates can quickly degrade. On the other hand, the recently proposed Heteroscedastic Probabilistic PCA Technique (HePPCAT) addresses this heterogeneity, but it was not designed to handle missing entries and streaming data, nor does it adapt to non-stationary behavior in time series data. This paper proposes the Streaming HeteroscedASTic Algorithm for PCA (SHASTA-PCA) to bridge this divide. SHASTA-PCA employs a stochastic alternating expectation maximization approach that jointly learns the low-rank latent factors and the unknown noise variances from streaming data that may have missing entries and heteroscedastic noise, all while maintaining a low memory and computational footprint. Numerical experiments validate the superior subspace estimation of our method compared to state-of-the-art streaming PCA algorithms in the heteroscedastic setting. Finally, we illustrate SHASTA-PCA applied to highly-heterogeneous real data from astronomy." @default.
- W4387560894 created "2023-10-12" @default.
- W4387560894 creator A5012041761 @default.
- W4387560894 creator A5027207271 @default.
- W4387560894 creator A5029521003 @default.
- W4387560894 creator A5063636172 @default.
- W4387560894 date "2023-10-09" @default.
- W4387560894 modified "2023-10-13" @default.
- W4387560894 title "Streaming Probabilistic PCA for Missing Data with Heteroscedastic Noise" @default.
- W4387560894 doi "https://doi.org/10.48550/arxiv.2310.06277" @default.
- W4387560894 hasPublicationYear "2023" @default.
- W4387560894 type Work @default.
- W4387560894 citedByCount "0" @default.
- W4387560894 crossrefType "posted-content" @default.
- W4387560894 hasAuthorship W4387560894A5012041761 @default.
- W4387560894 hasAuthorship W4387560894A5027207271 @default.
- W4387560894 hasAuthorship W4387560894A5029521003 @default.
- W4387560894 hasAuthorship W4387560894A5063636172 @default.
- W4387560894 hasBestOaLocation W43875608941 @default.
- W4387560894 hasConcept C101104100 @default.
- W4387560894 hasConcept C115961682 @default.
- W4387560894 hasConcept C119857082 @default.
- W4387560894 hasConcept C124101348 @default.
- W4387560894 hasConcept C154945302 @default.
- W4387560894 hasConcept C27438332 @default.
- W4387560894 hasConcept C32834561 @default.
- W4387560894 hasConcept C41008148 @default.
- W4387560894 hasConcept C49937458 @default.
- W4387560894 hasConcept C9357733 @default.
- W4387560894 hasConcept C99498987 @default.
- W4387560894 hasConceptScore W4387560894C101104100 @default.
- W4387560894 hasConceptScore W4387560894C115961682 @default.
- W4387560894 hasConceptScore W4387560894C119857082 @default.
- W4387560894 hasConceptScore W4387560894C124101348 @default.
- W4387560894 hasConceptScore W4387560894C154945302 @default.
- W4387560894 hasConceptScore W4387560894C27438332 @default.
- W4387560894 hasConceptScore W4387560894C32834561 @default.
- W4387560894 hasConceptScore W4387560894C41008148 @default.
- W4387560894 hasConceptScore W4387560894C49937458 @default.
- W4387560894 hasConceptScore W4387560894C9357733 @default.
- W4387560894 hasConceptScore W4387560894C99498987 @default.
- W4387560894 hasLocation W43875608941 @default.
- W4387560894 hasOpenAccess W4387560894 @default.
- W4387560894 hasPrimaryLocation W43875608941 @default.
- W4387560894 hasRelatedWork W2038165226 @default.
- W4387560894 hasRelatedWork W2052845382 @default.
- W4387560894 hasRelatedWork W2351648145 @default.
- W4387560894 hasRelatedWork W2410652950 @default.
- W4387560894 hasRelatedWork W3024870410 @default.
- W4387560894 hasRelatedWork W3148934225 @default.
- W4387560894 hasRelatedWork W4254549582 @default.
- W4387560894 hasRelatedWork W4283773154 @default.
- W4387560894 hasRelatedWork W4380150146 @default.
- W4387560894 hasRelatedWork W1676609285 @default.
- W4387560894 isParatext "false" @default.
- W4387560894 isRetracted "false" @default.
- W4387560894 workType "article" @default.