Robust Subspace Clustering via Thresholding

@article{Heckel2015RobustSC,
  title={Robust Subspace Clustering via Thresholding},
  author={Reinhard Heckel and Helmut B{\"o}lcskei},
  journal={IEEE Transactions on Information Theory},
  year={2015},
  volume={61},
  pages={6320-6342}
}
The problem of clustering noisy and incompletely observed high-dimensional data points into a union of low-dimensional subspaces and a set of outliers is considered. The number of subspaces, their dimensions, and their orientations are assumed unknown. We propose a simple low-complexity subspace clustering algorithm, which applies spectral clustering to an adjacency matrix obtained by thresholding the correlations between data points. In other words, the adjacency matrix is constructed from the… 

Figures from this paper

Subspace clustering via thresholding and spectral clustering
TLDR
A simple and low-complexity clustering algorithm based on thresholding the correlations between the data points followed by spectral clustering is proposed, which proves that this algorithm succeeds even when the subspaces intersect, and when the dimensions of the subSpaces scale linearly in the ambient dimension.
Noisy subspace clustering via thresholding
TLDR
This work considers the problem of clustering noisy high-dimensional data points into a union of low-dimensional subspaces and a set of outliers and finds that the simple outlier detection scheme introduced in [1] provably succeeds in the noisy case.
Greedy Subspace Clustering
TLDR
The statistical analysis shows that the algorithms are guaranteed exact (perfect) clustering performance under certain conditions on the number of points and the affinity between subspaces, which are weaker than those considered in the standard statistical literature.
Subspace Clustering using Ensembles of $K$-Subspaces
TLDR
This work presents a novel geometric approach to the subspace clustering problem that leverages ensembles of the K-subspaces algorithm via the evidence accumulation clustering framework and proves general recovery guarantees for any algorithm that forms an affinity matrix with entries close to a monotonic transformation of pairwise absolute inner products.
Dimensionality-reduced subspace clustering
TLDR
It is found that dimensionality reduction down to the order of the subspace dimensions is possible without incurring significant performance degradation, and these results are order-wise optimal in the sense that reducing the dimensionality further leads to a fundamentally ill-posed clustering problem.
Dimensionality-reduced subspace clustering
TLDR
It is found that dimensionality reduction down to the order of the subspace dimensions is possible without incurring significant performance degradation, and these results are order-wise optimal in the sense that reducing the dimensionality further leads to a fundamentally ill-posed clustering problem.
A Theoretical Analysis of Noisy Sparse Subspace Clustering on Dimensionality-Reduced Data
TLDR
This paper study's analysis applies to the most general fully deterministic model, where both underlying subspaces and data points within each subspace are deterministically positioned, and also a wide range of dimensionality reduction techniques that fall into a subspace embedding framework.
Dimensionality Reduction for Sparse Subspace Clustering
TLDR
Analysis of performance guarantees for sparse subspace clustering (SSC) applied to data whose dimensionality was reduced using random projections shows that the dimensionality of the data can be reduced to the order of the dimensions of the subspaces without compromising the clustering performance, and reveals a tradeoff between the amount of dimensionality reduction tolerated and the affinities between the subSpaces.
Neighborhood selection for thresholding-based subspace clustering
TLDR
This paper proposes a variation of the recently introduced thresholding-based subspace clustering algorithm, which applies spectral clustering to an adjacency matrix constructed from the nearest neighbors of each data point with respect to the spherical distance measure.
Graph Connectivity in Noisy Sparse Subspace Clustering
TLDR
These results provide the first exact clustering guarantee of noisy SSC for subspaces of dimension greater then 3 and show that a simple post-processing procedure is capable of delivering consistent clustering under certain "general position" or "restricted eigenvalue" assumptions.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 68 REFERENCES
Subspace clustering via thresholding and spectral clustering
TLDR
A simple and low-complexity clustering algorithm based on thresholding the correlations between the data points followed by spectral clustering is proposed, which proves that this algorithm succeeds even when the subspaces intersect, and when the dimensions of the subSpaces scale linearly in the ambient dimension.
Noisy subspace clustering via thresholding
TLDR
This work considers the problem of clustering noisy high-dimensional data points into a union of low-dimensional subspaces and a set of outliers and finds that the simple outlier detection scheme introduced in [1] provably succeeds in the noisy case.
A Geometric Analysis of Subspace Clustering with Outliers
TLDR
A novel geometric analysis of an algorithm named sparse subspace clustering (SSC) is developed, which signicantly broadens the range of problems where it is provably eective and shows that SSC can recover multiple subspaces, each of dimension comparable to the ambient dimension.
A closed form solution to robust subspace estimation and clustering
TLDR
This work uses an augmented Lagrangian optimization framework, which requires a combination of the proposed polynomial thresholding operator with the more traditional shrinkage-thresholding operator, to solve the problem of fitting one or more subspace to a collection of data points drawn from the subspaces and corrupted by noise/outliers.
Neighborhood selection for thresholding-based subspace clustering
TLDR
This paper proposes a variation of the recently introduced thresholding-based subspace clustering algorithm, which applies spectral clustering to an adjacency matrix constructed from the nearest neighbors of each data point with respect to the spherical distance measure.
Sparse Subspace Clustering: Algorithm, Theory, and Applications
TLDR
This paper proposes and studies an algorithm, called sparse subspace clustering, to cluster data points that lie in a union of low-dimensional subspaces, and demonstrates the effectiveness of the proposed algorithm through experiments on synthetic data as well as the two real-world problems of motion segmentation and face clustering.
Sparse subspace clustering
TLDR
This work proposes a method based on sparse representation (SR) to cluster data drawn from multiple low-dimensional linear or affine subspaces embedded in a high-dimensional space and applies this method to the problem of segmenting multiple motions in video.
Robust Subspace Clustering
TLDR
This paper introduces an algorithm inspired by sparse subspace clustering (SSC) to cluster noisy data, and develops some novel theory demonstrating its correctness.
Greedy feature selection for subspace clustering
TLDR
This paper develops sufficient conditions for EFS with a greedy method for sparse signal recovery known as orthogonal matching pursuit (OMP) and provides an empirical study of feature selection strategies for signals living on unions of subspaces and characterize the gap between sparse recovery methods and nearest neighbor (NN)-based approaches.
Generalized principal component analysis (GPCA)
  • R. Vidal, Yi Ma, S. Sastry
  • Computer Science, Mathematics
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2005
TLDR
An algebro-geometric solution to the problem of segmenting an unknown number of subspaces of unknown and varying dimensions from sample data points and applications of GPCA to computer vision problems such as face clustering, temporal video segmentation, and 3D motion segmentation from point correspondences in multiple affine views are presented.
...
1
2
3
4
5
...