Supervised Quadratic Feature Analysis: An information geometry approach to dimensionality reduction
Published in arXiv, 2024
In this paper we propose a novel method for supervised dimensionality reduction that maximizes second-order separability between classes. We use a geometric approach, maximizing separability of second-order differences in the manifold of symmetric positive definite (SPD) matrices.
Recommended citation: Herrera-Esposito, D.; Burge, J (2025). "Supervised Quadratic Feature Analysis: An information geometry approach to dimensionality reduction." arXiv.
Download Paper | Data and code