Classification with the matrix-variate-$t$ distribution
This work addresses classification tasks in domains such as forensic science and medical imaging, but it appears incremental as it extends existing methods to matrix-variate distributions.
The paper tackles classification problems using matrix-variate t-distributions by developing an Expectation-Maximization algorithm, showing promise on simulated and real-world datasets like forensic matching and image classification.
Matrix-variate distributions can intuitively model the dependence structure of matrix-valued observations that arise in applications with multivariate time series, spatio-temporal or repeated measures. This paper develops an Expectation-Maximization algorithm for discriminant analysis and classification with matrix-variate $t$-distributions. The methodology shows promise on simulated datasets or when applied to the forensic matching of fractured surfaces or the classification of functional Magnetic Resonance, satellite or hand gestures images.