Reduced-Rank Local Distance Metric Learning for k-NN Classification
This work addresses metric learning for k-NN classification, which is incremental as it builds on existing local metric learning approaches with new algorithmic formulations.
The paper tackles the problem of improving k-NN classification by learning local distance metrics that make similar samples close and dissimilar samples far apart, using sample similarity as side information. Experimental results show the method exhibits notable performance advantages over recent alternative metric learning approaches.
We propose a new method for local distance metric learning based on sample similarity as side information. These local metrics, which utilize conical combinations of metric weight matrices, are learned from the pooled spatial characteristics of the data, as well as the similarity profiles between the pairs of samples, whose distances are measured. The main objective of our framework is to yield metrics, such that the resulting distances between similar samples are small and distances between dissimilar samples are above a certain threshold. For learning and inference purposes, we describe a transductive, as well as an inductive algorithm; the former approach naturally befits our framework, while the latter one is provided in the interest of faster learning. Experimental results on a collection of classification problems imply that the new methods may exhibit notable performance advantages over alternative metric learning approaches that have recently appeared in the literature.