用于多变量数据分类的局部级联集成（CS LG）---刘持诚

我们提出了 LCE，一种用于传统（表格式）多变量数据分类的局部级联集成（LCE）及其扩展的用于多变量时间序列（MTS）分类的 LCEM。LCE 是一种新的混合集成方法，它结合了显式升压-袋式方法来处理机器学习模型通常面临的偏置-方差平衡，以及隐式分而治之方法来对训练数据的不同部分进行分类器误差的个体化。我们的评估首先表明，混合集成方法 LCE 在 UCI 数据集上优于最先进的分类器 LCEM 在 UEA 数据集上优于最先进的 MTS 分类器。此外，LCEM 通过设计提供了可解释性，并且在面对连续数据收集所带来的挑战（不同的 MTS长度、缺失数据和噪声）时表现出了稳健的性能。

原文题目：Local Cascade Ensemble for Multivariate Data Classification

原文：We present LCE, a Local Cascade Ensemble for traditional (tabular) multivariate data classification, and its extension LCEM for Multivariate Time Series (MTS) classification. LCE is a new hybrid ensemble method that combines an explicit boosting-bagging approach to handle the usual bias-variance tradeoff faced by machine learning models and an implicit divide-and-conquer approach to individualize classifier errors on different parts of the training data. Our evaluation firstly shows that the hybrid ensemble method LCE outperforms the state-of-the-art classifiers on the UCI datasets and that LCEM outperforms the state-of-the-art MTS classifiers on the UEA datasets. Furthermore, LCEM provides explainability by design and manifests robust performance when faced with challenges arising from continuous data collection (different MTS length, missing data and noise).

原文作者：Kevin Fauvel, Élisa Fromont, Véronique Masson, Philippe Faverdin, Alexandre Termie

原文地址：https://arxiv.org/abs/2005.03645