Representative null space LDA for discriminative dimensionality reduction
Representative null space LDA for discriminative dimensionality reduction
复制标题
用于判别维数降低的代表性零空间 LDA
DOI：
10.1016/j.patcog.2020.107664
复制
发表时间：
2021-03-01
影响因子：
8
通讯作者：
Tan, Jianrong
中科院分区：
文献类型：
--
作者：
He, Zaixing;Wu, Mengtian;Tan, Jianrong
空间线性判别分析（NLDA）是二十多年前为了克服LDA在实际应用中的奇异性问题而提出的。经过二十多年的技术发展，人们提出了许多优于NLDA的鉴别性简化（DDR）方法。本文提供了新的见解NLDA和说明NLDA是更强大的解决其固有的问题后。NLDA的主要问题是固有的过拟合问题。提出了一种理想的NLDA模型来分析其过拟合问题。在理想NLDA模型的基础上，提出了一种更合理的防止过拟合的代表性NLDA（RNLDA）方法。提出了两种简单而有效的RNLDA算法来实现RNLDA方法，并给出了理论证明。本研究从理论上分析并指出，应用经典但简单的hold-out预训练方法可以自动设置唯一的参数，以达到高性能。八个数据库的广泛实验表明，上级性能的RNLDA方法比国家的最先进的DDR方法。(C)2020爱思唯尔有限公司保留所有权利。
Null space Linear Discriminant Analysis (NLDA) was proposed twenty years ago to overcome the singularity problem of LDA in practical applications. With two decades of technique development, many Discriminative Dimensionality Reduction (DDR) methods that outperform NLDA have been proposed. This paper provides new insight into NLDA and illustrates that NLDA is much more powerful after solving its inherent problem. The main problem of NLDA is the intrinsic overfitting problem. An ideal NLDA model is proposed to analyze its overfitting problem. Based on the ideal NLDA model, a more reasonable Representative NLDA (RNLDA) method is proposed to prevent overfitting. Two simple but efficient RNLDA algorithms are proposed to implement the RNLDA method with a theoretical proof. This study theoretically analyzed and indicated that applying the classical but simple hold-out pretraining method can automatically set the only parameter to achieve high performance. Extensive experiments with eight databases demonstrate the superior performance of the RNLDA method over state-of-the-art DDR methods. (C) 2020 Elsevier Ltd. All rights reserved.