Beyond CCA: Moment Matching for Multi-View Models

Abstract : We introduce three novel semi-parametric extensions of probabilistic canonical correlation analysis with identifiability guarantees. We consider moment matching techniques for estimation in these models. For that, by drawing explicit links between the new models and a discrete version of independent component analysis (DICA), we first extend the DICA cumulant tensors to the new discrete version of CCA. By further using a close connection with independent component analysis, we introduce generalized covariance matrices , which can replace the cumulant tensors in the moment matching framework, and, therefore, improve sample complexity and simplify derivations and algorithms significantly. As the tensor power method or orthogonal joint diagonalization are not applicable in the new setting, we use non-orthogonal joint diago-nalization techniques for matching the cumulants. We demonstrate performance of the proposed models and estimation techniques on experiments with both synthetic and real datasets.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [59 references]  Display  Hide  Download

https://hal.inria.fr/hal-01291060
Contributor : Anastasia Podosinnikova <>
Submitted on : Sunday, March 20, 2016 - 6:48:47 PM
Last modification on : Wednesday, January 30, 2019 - 11:07:38 AM
Long-term archiving on: Tuesday, June 21, 2016 - 10:13:43 AM

File

cca16.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01291060, version 1
  • ARXIV : 1602.09013

Collections

Citation

Anastasia Podosinnikova, Francis Bach, Simon Lacoste-Julien. Beyond CCA: Moment Matching for Multi-View Models. 2016. ⟨hal-01291060⟩

Share

Metrics

Record views

269

Files downloads

144