Benchmarking Manifold Learning Methods on a Large Collection of Datasets

Manifold learning, a non-linear approach of dimensionality reduction, assumes that the dimensionality of multiple datasets is artificially high and a reduced number of dimensions is sufficient to maintain the information about the data. In this paper, a large scale comparison of manifold learning techniques is performed for the task of classification. We show the current standing of genetic programming (GP) for the task of classification by comparing the classification results of two GP-based manifold leaning methods: GP-Mal and ManiGP - an experimental manifold learning technique proposed in this paper. We show that GP-based methods can more effectively learn a manifold across a set of 155 different problems and deliver more separable embeddings than many established methods.