Supervised Class-pairwise NMF for Data Representation and Classification

TL;DR

This paper introduces a supervised, evolutionary approach to optimize parameterized NMF for improved classification by applying NMF to class pairs and using genetic algorithms for hyper-parameter tuning.

Contribution

It proposes a novel supervised framework that learns hyper-parameters and applies NMF to class pairs, enhancing classification performance over traditional unsupervised NMF.

Findings

Improved classification accuracy on real datasets.

Effective hyper-parameter optimization via genetic algorithms.

Supervised class-pairwise NMF outperforms traditional methods.

Abstract

Various Non-negative Matrix factorization (NMF) based methods add new terms to the cost function to adapt the model to specific tasks, such as clustering, or to preserve some structural properties in the reduced space (e.g., local invariance). The added term is mainly weighted by a hyper-parameter to control the balance of the overall formula to guide the optimization process towards the objective. The result is a parameterized NMF method. However, NMF method adopts unsupervised approaches to estimate the factorizing matrices. Thus, the ability to perform prediction (e.g. classification) using the new obtained features is not guaranteed. The objective of this work is to design an evolutionary framework to learn the hyper-parameter of the parameterized NMF and estimate the factorizing matrices in a supervised way to be more suitable for classification problems. Moreover, we claim that applying NMF-based algorithms separately to different class-pairs instead of applying it once to the whole dataset improves the effectiveness of the matrix factorization process. This results in training multiple parameterized NMF algorithms with different balancing parameter values. A cross-validation combination learning framework is adopted and a Genetic Algorithm is used to identify the optimal set of hyper-parameter values. The experiments we conducted on both real and synthetic datasets demonstrated the effectiveness of the proposed approach.