Distributed classification based on distances between probability distributions in feature space

作     者：Montero-Manso, Pablo Moran-Fernandez, Laura Bolon-Canedo, Veronica Vilar, Jose A. Alonso-Betanzos, Amparo 

作者机构：Univ A Coruna Res Grp Modeling Optimizat & Stat Inference MODES Dept Math Comp Sci Fac La Coruna Spain Univ A Coruna Lab Res & Dev Artificial Intelligence LIDIA Dept Comp Sci Comp Sci Fac La Coruna Spain 

出 版 物：《INFORMATION SCIENCES》 (信息科学)

年 卷 期：2019年第496卷

页      面：431-450页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Spanish Ministerio de Economia y Competitividad - European Regional Development Fund, ERDF [TIN2015-65069-C2-1-R, MTM2014-52876-R, MTM2017-82724-R] Conselleria de Industria of the Xunta de Galicia - European Regional Development Fund, ERDF [GRC2014/035] Conselleria de Industria of the Xunta de Galicia (Grupos de Referencia Competitiva) - European Regional Development Fund, ERDF [ED431D-R2016/045, ED431C-2016-015] Conselleria de Industria of the Xunta de Galicia (Centro Singular de Investigacion de Galicia) - European Regional Development Fund, ERDF [ED431G/01] 

主　　题：Distributed classification Distributional distances Classifiers combination Imbalanced data set Classification accuracy 

摘      要：We consider a distributed framework where training and test samples drawn from the same distribution are available, with the training instances spread across disjoint nodes. In this setting, a novel learning algorithm based on combining with different weights the outputs of classifiers trained at each node is proposed. The weights depend on the distributional distance between each node and the test set in the feature space. Two different weighting approaches are introduced, which are referred to as per-Node Weighting (pNW) and per-Instance Weighting (pIW). While pNW assigns the same weight to all test instances at each node, pIW allows distinct weights for test instances differently represented at the node. By construction, our approach is particularly useful to deal with unbalanced nodes. Our methods require no communication between nodes, allowing for data privacy, independence of the kind of trained classifier at each node and maximum training speedup. In fact, our methods do not require retraining of the node s classifiers if available. Although a range of different combination rules are considered to ensemble the single classifiers, theoretical support for the optimality of using the sum rule is provided. Our experiments illustrate all of these properties and show that pIW produces the highest classification accuracies compared with pNW and the standard unweighted approaches. (C) 2019 Elsevier Inc. All rights reserved.