Open EvertBuzonBadillo opened 4 months ago
For Gaussian Mixture the general setup would be nice to see (eg. latent variables, distributional assumptions) and maybe definition of aleatoric and epistemic uncertainity in outlier detection context at beginning.
Maybe split the plot KDE/GaussMix plot into two. One for the density esimate vs variable and one for accuracy. Or write a short explanation of the plots and their interpretation.
maybe use bold case notation for vector as in paper instead of arrow notation?