Accession Number:

ADA588768

Title:

Adaptive Kernel Based Machine Learning Methods

Descriptive Note:

Final rept. 1 Jul 2009-30 Jun 2012

Corporate Author:

SYRACUSE UNIV NY

Personal Author(s):

Report Date:

2012-10-15

Pagination or Media Count:

5.0

Abstract:

Research results obtained from this project address the kernel selection problem in machine learning. Specifically, motivated from the need of updating the current operator-valued reproducing kernel in multi-task learning when underfitting or overfitting occurs, we studied the construction of a refinement kernel for a given operator-valued reproducing kernel such that the vector-valued reproducing kernel Hilbert space of the refinement kernel contains that of the given kernel as a subspace. We also developed a complete characterization of multi-task finite rank kernels in terms of the positivity of what we call its associated characteristic operator

Subject Categories:

  • Cybernetics

Distribution Statement:

APPROVED FOR PUBLIC RELEASE