Abstract | ||
---|---|---|
Sparse non-Gaussian component analysis is an unsupervised linear method of extracting any structure from high-dimensional distributed data based on estimating a low-dimensional non-Gaussian data component. In this paper we discuss a new approach with known apriori reduced dimension to direct estimation of the projector on the target space using semidefinite programming. The new approach avoids the estimation of the data covariance matrix and overcomes the traditional separation of element estimation of the target space and target space reconstruction. This allows to reduced the sampling size while improving the sensitivity to a broad variety of deviations from normality. Moreover the complexity of the new approach is limited to O(dlogd). We also discuss the procedures which allows to recover the structure when its effective dimension is unknown. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1007/s10994-013-5331-1 | Machine Learning |
Keywords | Field | DocType |
Dimension reduction,Non-Gaussian components analysis,Feature extraction | Effective dimension,Sparse PCA,Dimensionality reduction,Pattern recognition,A priori and a posteriori,Gaussian,Artificial intelligence,Component analysis,Semidefinite embedding,Mathematics,Machine learning,Semidefinite programming | Journal |
Volume | Issue | ISSN |
91 | 2 | 0885-6125 |
Citations | PageRank | References |
5 | 0.69 | 11 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Elmar Diederichs | 1 | 9 | 1.97 |
Anatoli Juditsky | 2 | 663 | 72.50 |
Arkadi Nemirovski | 3 | 1642 | 186.22 |
Vladimir Spokoiny | 4 | 90 | 10.53 |