Multi-view clustering via canonical correlation analysis

@inproceedings{Chaudhuri2009MultiviewCV,
  title={Multi-view clustering via canonical correlation analysis},
  author={Kamalika Chaudhuri and Sham M. Kakade and Karen Livescu and Karthik Sridharan},
  booktitle={ICML '09},
  year={2009}
}
Clustering data in high dimensions is believed to be a hard problem in general. A number of efficient clustering algorithms developed in recent years address this problem by projecting the data into a lower-dimensional subspace, e.g. via Principal Components Analysis (PCA) or random projections, before clustering. Here, we consider constructing such projections using multiple views of the data, via Canonical Correlation Analysis (CCA). Under the assumption that the views are un-correlated… 

Figures and Tables from this paper

Co-regularized kernel k-means for multi-view clustering

TLDR
This work proposes a novel clustering algorithm where the intrinsic clustering is found by maximizing the sum of weighted similarities between clusterings of different views, along with a three-step alternative algorithm designed to solve the problem efficiently.

Multiple-View Spectral Embedded Clustering Using a Co-training Approach

TLDR
This paper derives a co-training algorithm to obtain a most informative clustering by iteratively modifying the affinity graph used for one view using the discriminative information from the other views by spectral embedded clustering which performs well on high-dimensional data without a clear low-dimensional manifold structure.

Multi-view clustering via pairwise sparse subspace representation

Co-regularized Multi-view Spectral Clustering

TLDR
A spectral clustering framework is proposed that achieves this goal by co-regularizing the clustering hypotheses, and two co- regularization schemes are proposed to accomplish this.

A subspace co-training framework for multi-view clustering

Incomplete Multi-view Clustering via Subspace Learning

TLDR
A novel incompletemulti-view clustering method is developed, which learns unified latent representations and projection matrices for the incomplete multi-view data and performs better than the state-of-the-art multi-View clustering methods in various settings.

Manifold Regularized Multi-view Subspace Clustering for image representation

TLDR
This paper presents a Manifold Regularized Multi-view Subspace Clustering (MRMSC) method and shows that better clustering performance can be achieved by fusing the multiple features with a unified affinity representation by MRMSC.

Multi-view spectral clustering via robust local subspace learning

TLDR
This work proposed a novel multi-view clustering algorithm via robust local representation that simultaneously take all the local reconstruction relationships into consideration and use L1-norm to guarantee the sparsity.
...

References

SHOWING 1-10 OF 24 REFERENCES

Multi-view Regression Via Canonical Correlation Analysis

TLDR
This work provides a semi-supervised algorithm which first uses unlabeled data to learn a norm (or, equivalently, a kernel) and then uses labeled data in a ridge regression algorithm (with this induced norm) to provide the predictor.

Correlational spectral clustering

TLDR
The proposed method uses separate similarity measures for each data representation, and allows for projection of previously unseen data that are only observed in one representation (e.g. images but not text).

Isotropic PCA and Affine-Invariant Clustering

  • S. BrubakerS. Vempala
  • Computer Science, Mathematics
    2008 49th Annual IEEE Symposium on Foundations of Computer Science
  • 2008
TLDR
An extension of principal component analysis (PCA) and a new algorithm for clustering points in \Rn based on it that is affine-invariant and nearly the best possible is presented, improving known results substantially.

On Spectral Learning of Mixtures of Distributions

TLDR
It is proved that a very simple algorithm, namely spectral projection followed by single-linkage clustering, properly classifies every point in the sample, and there are many Gaussian mixtures such that each pair of means is separated, yet upon spectral projection the mixture collapses completely.

LEARNING MIXTURES OF SEPARATED NONSPHERICAL GAUSSIANS

TLDR
This work presents the first algorithm that provably learns the component Gaussians in time that is polynomial in the dimension, and formalizes the more general problem of max-likelihood fit of a Gaussian mixture to unstructured data.

The Spectral Method for General Mixture Models

TLDR
A general property of spectral projection for arbitrary mixtures is proved and it is shown that the resulting algorithm is efficient when the components of the mixture are logconcave distributions in R n whose means are separated.

Canonical Correlation Analysis: An Overview with Application to Learning Methods

TLDR
A general method using kernel canonical correlation analysis to learn a semantic representation to web images and their associated text and compares orthogonalization approaches against a standard cross-representation retrieval technique known as the generalized vector space model is presented.

Combining labeled and unlabeled data with co-training

TLDR
A PAC-style analysis is provided for a problem setting motivated by the task of learning to classify web pages, in which the description of each example can be partitioned into two distinct views, to allow inexpensive unlabeled data to augment, a much smaller set of labeled examples.

Two-view feature generation model for semi-supervised learning

TLDR
The two-view feature generation model of co-training is revisited and it is proved that the optimum predictor can be expressed as a linear combination of a few features constructed from unlabeled data.

A spectral algorithm for learning mixtures of distributions

  • S. VempalaGrant J. Wang
  • Computer Science
    The 43rd Annual IEEE Symposium on Foundations of Computer Science, 2002. Proceedings.
  • 2002
We show that a simple spectral algorithm for learning a mixture of k spherical Gaussians in /spl Ropf//sup n/ works remarkably well - it succeeds in identifying the Gaussians assuming essentially the