Our goal is to develop statistical methods over the space of phylogenetic trees, which coincides with the ultrametric space and conforms to spaces from tropical geometry. I have developed a\u00a0 “tropical” version of logistic regression, which classifies sampled gene trees into two classes characterized by a species tree, which is treated as a statistical parameter to be inferred. The model is easily generalised to a tropical version of generalised linear models, which can be stacked to create a “tropical” neural network. Using simulated gene trees, we have shown that our models are significantly better at classification than classical models.<\/span><\/p>\u00a0<\/span><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t