Automated Model Selection for Generalized Linear Models

arXiv:2404.16560v1 Announce Type: cross
Abstract: In this paper, we show how mixed-integer conic optimization can be used to combine feature subset selection with holistic generalized linear models to fully automate the model selection process. Concretely, we directly optimize for the Akaike and Bayesian information criteria while imposing constraints designed to deal with multicollinearity in the feature selection task. Specifically, we propose a novel pairwise correlation constraint that combines the sign coherence constraint with ideas from classical statistical models like Ridge …

abstract arxiv automate automated bayesian constraints cs.lg deal feature generalized information integer linear math.oc mixed model selection multicollinearity optimization paper process show stat.ml type while