Expectation-maximization for sparse and non-negative PCA

Authors:
Christian D. Sigg;Joachim M. Buhmann
Affiliations:
ETH Zurich, Zurich, Switzerland;ETH Zurich, Zurich, Switzerland
Venue:
Proceedings of the 25th international conference on Machine learning
Year:
2008

Citing 7
Cited 5

Mixtures of probabilistic principal component analyzers

Neural Computation
Learning and example selection for object and pattern detection

Learning and example selection for object and pattern detection
Novel Unsupervised Feature Filtering of Biological Data

Bioinformatics
Introduction to Global Optimization (Nonconvex Optimization and Its Applications)

Introduction to Global Optimization (Nonconvex Optimization and Its Applications)
Multiplicative Updates for Nonnegative Quadratic Programming

Neural Computation
Full regularization path for sparse principal component analysis

Proceedings of the 24th international conference on Machine learning
Sparse eigen methods by D.C. programming

Proceedings of the 24th international conference on Machine learning

Discovering Sparse Functional Brain Networks Using Group Replicator Dynamics (GRD)

IPMI '09 Proceedings of the 21st International Conference on Information Processing in Medical Imaging
Multi-subject dictionary learning to segment an atlas of brain spontaneous activity

IPMI'11 Proceedings of the 22nd international conference on Information processing in medical imaging
Improve robustness of sparse PCA by L1-norm maximization

Pattern Recognition
Learning dictionary from signals under global sparsity constraint

Neurocomputing
Editorial: Following the entire solution path of sparse principal component analysis by coordinate-pairwise algorithm

Data & Knowledge Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

We study the problem of finding the dominant eigenvector of the sample covariance matrix, under additional constraints on the vector: a cardinality constraint limits the number of non-zero elements, and non-negativity forces the elements to have equal sign. This problem is known as sparse and non-negative principal component analysis (PCA), and has many applications including dimensionality reduction and feature selection. Based on expectation-maximization for probabilistic PCA, we present an algorithm for any combination of these constraints. Its complexity is at most quadratic in the number of dimensions of the data. We demonstrate significant improvements in performance and computational efficiency compared to other constrained PCA algorithms, on large data sets from biology and computer vision. Finally, we show the usefulness of non-negative sparse PCA for unsupervised feature selection in a gene clustering task.