{"title":"最优混合整数线性优化训练的多元分类树","authors":"Brandon Alston, Illya V. Hicks","doi":"arxiv-2408.01297","DOIUrl":null,"url":null,"abstract":"Multivariate decision trees are powerful machine learning tools for\nclassification and regression that attract many researchers and industry\nprofessionals. An optimal binary tree has two types of vertices, (i) branching\nvertices which have exactly two children and where datapoints are assessed on a\nset of discrete features and (ii) leaf vertices at which datapoints are given a\nprediction, and can be obtained by solving a biobjective optimization problem\nthat seeks to (i) maximize the number of correctly classified datapoints and\n(ii) minimize the number of branching vertices. Branching vertices are linear\ncombinations of training features and therefore can be thought of as\nhyperplanes. In this paper, we propose two cut-based mixed integer linear\noptimization (MILO) formulations for designing optimal binary classification\ntrees (leaf vertices assign discrete classes). Our models leverage on-the-fly\nidentification of minimal infeasible subsystems (MISs) from which we derive\ncutting planes that hold the form of packing constraints. We show theoretical\nimprovements on the strongest flow-based MILO formulation currently in the\nliterature and conduct experiments on publicly available datasets to show our\nmodels' ability to scale, strength against traditional branch and bound\napproaches, and robustness in out-of-sample test performance. Our code and data\nare available on GitHub.","PeriodicalId":501216,"journal":{"name":"arXiv - CS - Discrete Mathematics","volume":"47 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Optimal Mixed Integer Linear Optimization Trained Multivariate Classification Trees\",\"authors\":\"Brandon Alston, Illya V. Hicks\",\"doi\":\"arxiv-2408.01297\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Multivariate decision trees are powerful machine learning tools for\\nclassification and regression that attract many researchers and industry\\nprofessionals. An optimal binary tree has two types of vertices, (i) branching\\nvertices which have exactly two children and where datapoints are assessed on a\\nset of discrete features and (ii) leaf vertices at which datapoints are given a\\nprediction, and can be obtained by solving a biobjective optimization problem\\nthat seeks to (i) maximize the number of correctly classified datapoints and\\n(ii) minimize the number of branching vertices. Branching vertices are linear\\ncombinations of training features and therefore can be thought of as\\nhyperplanes. In this paper, we propose two cut-based mixed integer linear\\noptimization (MILO) formulations for designing optimal binary classification\\ntrees (leaf vertices assign discrete classes). Our models leverage on-the-fly\\nidentification of minimal infeasible subsystems (MISs) from which we derive\\ncutting planes that hold the form of packing constraints. We show theoretical\\nimprovements on the strongest flow-based MILO formulation currently in the\\nliterature and conduct experiments on publicly available datasets to show our\\nmodels' ability to scale, strength against traditional branch and bound\\napproaches, and robustness in out-of-sample test performance. Our code and data\\nare available on GitHub.\",\"PeriodicalId\":501216,\"journal\":{\"name\":\"arXiv - CS - Discrete Mathematics\",\"volume\":\"47 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-08-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Discrete Mathematics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2408.01297\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Discrete Mathematics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2408.01297","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Optimal Mixed Integer Linear Optimization Trained Multivariate Classification Trees
Multivariate decision trees are powerful machine learning tools for
classification and regression that attract many researchers and industry
professionals. An optimal binary tree has two types of vertices, (i) branching
vertices which have exactly two children and where datapoints are assessed on a
set of discrete features and (ii) leaf vertices at which datapoints are given a
prediction, and can be obtained by solving a biobjective optimization problem
that seeks to (i) maximize the number of correctly classified datapoints and
(ii) minimize the number of branching vertices. Branching vertices are linear
combinations of training features and therefore can be thought of as
hyperplanes. In this paper, we propose two cut-based mixed integer linear
optimization (MILO) formulations for designing optimal binary classification
trees (leaf vertices assign discrete classes). Our models leverage on-the-fly
identification of minimal infeasible subsystems (MISs) from which we derive
cutting planes that hold the form of packing constraints. We show theoretical
improvements on the strongest flow-based MILO formulation currently in the
literature and conduct experiments on publicly available datasets to show our
models' ability to scale, strength against traditional branch and bound
approaches, and robustness in out-of-sample test performance. Our code and data
are available on GitHub.