This repository contains Matlab implementations of the algorithms proposed in "AUC Maximization in Bayesian Hierarchical Models", which is appearing in the Proceedings of the 22nd European Conference on Artificial Intelligence (ECAI 2016). Baseline methods ---------------- * bprobit_classification_variational_train.m => training procedure for Bayesian linear classification * bprobit_classification_variational_test.m => test procedure for Bayesian linear classification BAM methods ----------- * bam_pairwise_classification_variational_train.m => training procedure for Bayesian AUC maximization using pairwise ranking * bam_pairwise_classification_variational_test.m => test procedure for Bayesian AUC maximization using pairwise ranking * bam_listwise_classification_variational_train.R => training procedure for Bayesian AUC maximization using listwise ranking * bam_listwise_classification_variational_test.R => test procedure for Bayesian AUC maximization using listwise ranking If you use any of the algorithms implemented in this repository, please cite the following paper: Mehmet Gonen. AUC Maximization in Bayesian Hierarchical Models. Proceedings of the 22nd European Conference on Artificial Intelligence (ECAI 2016), The Hague, The Netherlands, 2016.