Abstract:For many machine learning algorithms such as k-nearest neighbor ( k-NN) classifiers and k-means clustering, often their success heavily depends on the metric used to calculate distances between different data points. An effective solution for defining such a metric is to learn it from a set of labeled training samples. In this work, we propose a fast and scalable algorithm to learn a Mahalanobis distance metric. The Mahalanobis metric can be viewed as the Euclidean distance metric on the input data that have been linearly transformed. By employing the principle of margin maximization to achieve better generalization performances, this algorithm formulates the metric learning as a convex optimization problem and a positive semidefinite (p.s.d.) matrix is the unknown variable. Based on an important theorem that a p.s.d. trace-one matrix can always be represented as a convex combination of multiple rank-one matrices, our algorithm accommodates any differentiable loss function and solves the resulting optimization problem using a specialized gradient descent procedure. During the course of optimization, the proposed algorithm maintains the positive semidefiniteness of the matrix variable that is essential for a Mahalanobis metric. Compared with conventional methods like standard interior-point algorithms or the special solver used in large margin nearest neighbor , our algorithm is much more efficient and has a better performance in scalability. Experiments on benchmark data sets suggest that, compared with state-of-the-art metric learning algorithms, our algorithm can achieve a comparable classification accuracy with reduced computational complexity.

Data Splitting Method of Distance Metric Learning Based on Gaussian Mixed Model

Gaussian mixture density modeling and decomposition with weighted likelihood

A revision for gaussian mixture density decomposition algorithm

A Method of Dividing Clinical Data Set for Medical Image AI Training

A Distributed Approach Toward Discriminative Distance Metric Learning

Improve distance metric learning by learning positions of class centers

Distance-Dependent Metric Learning.

Metric learning by discriminant neighborhood embedding

A Boosting-Based Deep Distance Metric Learning Method

A Regularized Approach for Geodesic-Based Semisupervised Multimanifold Learning

Gaussian Mixture Model and Double-Weighted Deep Neural Networks for Data Augmentation Soft Sensing

Scalable Large-Margin Mahalanobis Distance Metric Learning

A Lie group based Gaussian Mixture Model distance measure for multimedia comparison.

Gaussian Mixture Distribution Makes Data Uncertainty Learning Better

Mixed data Deep Gaussian Mixture Model: A clustering model for mixed datasets

Regularized Large Margin Distance Metric Learning.

Multi-distance Metric Network for Few-Shot Learning

Towards Making High Dimensional Distance Metric Learning Practical

EMD Metric Learning

The Breakdown of Gaussian Universality in Classification of High-dimensional Mixtures

Metric Learning on Healthcare Data with Incomplete Modalities.