Star-Graph Multimodal Matching Component Analysis for Data Fusion and Transfer Learning

Previous matching component analysis (MCA) techniques map two data domains to a common domain for further processing in data fusion and transfer learning contexts. In this paper, we extend these techniques to the star-graph multimodal (SGM) case in which one particular data domain is connected to $m$ others via an objective function. We provide a particular feasible point for the resulting trace maximization problem in closed form and algorithms for its computation and iterative improvement, leading to our main result, the SGM maps. We also provide numerical examples demonstrating that SGM is capable of encoding into its maps more information than MCA when few training points are available. In addition, we develop a further generalization of the MCA covariance constraint, eliminating a previous feasibility condition and allowing larger values of the rank of the prescribed covariance matrix.

Covariance-Generalized Matching Component Analysis for Data Fusion and Transfer Learning

In order to allow for the encoding of additional statistical information in data fusion and transfer learning applications, we introduce a generalized covariance constraint for the matching component analysis (MCA) transfer learning technique. After proving a semi-orthogonally constrained trace maximization lemma, we develop a closed-form solution to the resulting covariance-generalized optimization problem and provide an algorithm for its computation. We call this technique -- applicable to both data fusion and transfer learning -- covariance-generalized MCA (CGMCA).