2 problems

0 votes0 replies0 views
Polynomial-time recovery conjecture for reweighted PCA in non-Gaussian component analysis
Let a random vector have a non-Gaussian subspace whose marginals in every non-Gaussian direction are far from Gaussian in terms of moments, and consider the Reweighted PCA algorith…
Conjectureopendimension-reduction non-gaussian-component-analysis spectral-algorithmsarXiv posted 2 days ago
0 votes0 replies2 views
Conjecture that reweighted PCA recovers the entire non-Gaussian component subspace
Let $E$ be the maximal low-dimensional subspace of $\mathbb{R}^n$ such that the projection of the data distribution onto $E$ is non-Gaussian, and let the Reweighted PCA algorithm b…
Conjectureopennon-gaussian-component-analysis sample-complexity spectral-methodsarXiv posted 2 days ago

Polynomial-time recovery conjecture for reweighted PCA in non-Gaussian component analysis