Data-Driven Diverse Ensembles of Logistic Regression Models

arXiv (Cornell University)(2021)

引用 0|浏览1
暂无评分
摘要
A novel framework for statistical learning is introduced which combines ideas from regularization and ensembling. This framework is applied to learn an ensemble of logistic regression models for high-dimensional binary classification. In the new framework the models in the ensemble are learned simultaneously by optimizing a multi-convex objective function. To enforce diversity between the models the objective function penalizes overlap between the models in the ensemble. Measures of diversity in classifier ensembles are used to show how our method learns the ensemble by exploiting the accuracy-diversity trade-off for ensemble models. In contrast to other ensembling approaches, the resulting ensemble model is fully interpretable as a logistic regression model, asymptotically consistent, and at the same time yields excellent prediction accuracy as demonstrated in an extensive simulation study and gene expression data applications. The models found by the proposed ensemble methodology can also reveal alternative mechanisms that can explain the relationship between the predictors and the response variable. An open-source compiled software library implementing the proposed method is briefly discussed.
更多
查看译文
关键词
diverse ensembles,logistic regression models,data-driven
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要