Deep Least Squares Fisher Discriminant Analysis
Deep Least Squares Fisher Discriminant Analysis
复制标题
DOI：
10.1109/tnnls.2019.2906302
复制
发表时间：
2020-08-01
影响因子：
10.4
通讯作者：
Dorronsoro, Jose R.
中科院分区：
文献类型：
--
作者：
Diaz-Vico, David;Dorronsoro, Jose R.
虽然 Fisher 线性判别分析 (FLDA) 是最早也是最优雅的降维工具之一，但目前并不被认为是特征提取或分类的顶级方法之一。在本文中，我们将回顾两种最新的 FLDA 方法，即最小二乘 Fisher 判别分析 (LSFDA) 和正则化核 FDA (RKFDA)，并提出深度 FDA (DFDA)，这是 LSFDA 的一种简单的非线性扩展，它利用了深度神经网络的最新进展。我们将比较 RKFDA 和 DFDA 在大量二类和多类问题上的性能，其中许多涉及类不平衡的数据集，有些样本量相当大；为此，我们将使用所考虑的分类器的接收者操作特征（ROC）曲线下的面积。正如我们将看到的，两种方法的分类性能通常非常相似，并且在处理不平衡问题时特别好，但构建 DFDA 模型比 RKFDA 模型快得多，特别是在样本量相当大的问题上。
While being one of the first and most elegant tools for dimensionality reduction, Fisher linear discriminant analysis (FLDA) is not currently considered among the top methods for feature extraction or classification. In this paper, we will review two recent approaches to FLDA, namely, least squares Fisher discriminant analysis (LSFDA) and regularized kernel FDA (RKFDA) and propose deep FDA (DFDA), a straightforward nonlinear extension of LSFDA that takes advantage of the recent advances on deep neural networks. We will compare the performance of RKFDA and DFDA on a large number of two-class and multiclass problems, many of them involving class-imbalanced data sets and some having quite large sample sizes; we will use, for this, the areas under the receiver operating characteristics (ROCs) curve of the classifiers considered. As we shall see, the classification performance of both methods is often very similar and particularly good on imbalanced problems, but building DFDA models is considerably much faster than doing so for RKFDA, particularly in problems with quite large sample sizes.