Low-rank Approximation of Linear Maps

Heas, Patrick, Herzet, Cedric

arXiv.org Machine Learning 

This work provides closed-form solutions and minimal achie vable errors for a large class of low-rank approximation problems in Hilbert spaces . The proposed theorem generalizes to the case of linear bounded operators andp-th Schatten norms previous results obtained in the finite dimensional case for the Frobenius norm. The theorem is illu strated in various settings, including low-rank approximation problems with respect to the trace n orm, the 2-induced norm or the Hilbert-Schmidt norm. The theorem provides also the basics for the de sign of tractable algorithms for kernel-based or continuous DMD.

Duplicate Docs Excel Report

Title
None found

Similar Docs  Excel Report  more

TitleSimilaritySource
None found