A Fast Logdet Divergence Based Metric Learning Algorithm for Large Data Sets Classification

Abstract EN

Large data sets classification is widely used in many industrial applications.

It is a challenging task to classifylarge data sets efficiently, accurately, and robustly, as large data sets always contain numerous instances with highdimensional feature space.

In order to deal with this problem, in this paper we present an online Logdet divergencebased metric learning (LDML) model by making use of the powerfulness of metric learning.

We firstly generatea Mahalanobis matrix via learning the training data with LDML model.

Meanwhile, we propose a compressedrepresentation for high dimensional Mahalanobis matrix to reduce the computation complexity in each iteration.

The final Mahalanobis matrix obtained this way measures the distances between instances accurately and serves asthe basis of classifiers, for example, the k-nearest neighbors classifier.

Experiments on benchmark data sets demonstrate that the proposed algorithm compares favorably with the state-of-the-art methods.