You are here

Reduced-Rank Local Distance Metric Learning

TitleReduced-Rank Local Distance Metric Learning
Publication TypeConference Paper
Year of Publication2013
AuthorsHuang Y, Li C, Georgiopoulos M, Anagnostopoulos GC
Conference NameMachine Learning and Knowledge Discovery in Databases - ECML/PKDD 2013
Conference LocationPrague, Czech Republic

We propose a new method for local metric learning based on a conical combination of Mahalanobis metrics and pair-wise similarities between the data. Its formulation allows for controlling the rank of the metrics’ weight matrices. We also offer a convergent algorithm for training the associated model. Experimental results on a collection of classification problems imply that the new method may offer notable performance advantages over alternative metric learning approaches that have recently appeared in the literature.


Acceptance rate 25% (111/443).


Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer