"Generalizing Distance Metric Learning (DML) using Evolutionary algorithms"
In this work, we propose a novel method for a supervised dimensionality reduction, which learns weights of a neural network using an evolutionaryalgorithm, CMA-ES, optimising the success rate of the k-NN classifier. If no activation functions are used in the neural network, the algorithm essentially performs a linear transformation, which can also be used inside of the Mahalanobis distance. Therefore our method can be considered to be a metric learning algorithm. By adding activations to the neural network, the algorithm can learn non-linear transformations as well. We consider reductions to low-dimensional spaces, which are useful for data visualisation, and demonstrate that the resulting projections provide better performance than other dimensionality reduction techniques and also that the visualisations provide better distinctions between the classes in the data thanks to the locality of the k-NN classifier.
Download the compiled pdf from Releases or by clicking here.