期刊
PATTERN RECOGNITION
卷 111, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.patcog.2020.107673
关键词
Supervised dimensionality reduction; Local structured feature learning; l(0)-Norm constraint optimization; Dynamic maximum entropy graph
资金
- National Key Research and Development Program of China [2018AAA0101902]
- National Natural Science Foundation of China [61936014, 61772427, 61751202]
- Fundamental Research Funds for the Central Universities [G2019KY0501]
This study introduces a novel method of local structured feature learning, combining Maximum Entropy Graph and Linear Discriminant Analysis, addressing the challenges LDA faces in handling multimodal data and being affected by noisy features. By simultaneously learning similarity and projection matrices, employing Maximum Entropy regularization, and utilizing an efficient iterative optimization algorithm, the proposed method enhances classification efficiency and robustness to noise.
In recent years, Linear Discriminant Analysis (LDA) has seen huge adoption in data mining applications. Due to its globality, it is incompetent to handle multimodal data. Besides, most of LDA's variants learn the projection matrix based on the pre-defined similarity matrix, which is easily affected by noisy and irrelevant features. To address above two issues, a novel local structured feature learning with Dynamic Maximum Entropy Graph (DMEG) method is developed which firstly develops a more discriminative LDA with whitening constraint that can minimize the within-class scatter while keeping the total samples scatter unchanged simultaneously. Second, for exploring the local structure of data, the l(0)-norm constraint is imposed on similarity matrix to ensure the k connectivity on graph. More importantly, proposed model learns the similarity and projection matrix simultaneously to ensure that the neighborships can be found in the optimal subspace where the noise have been removed already. Moreover, a maximum entropy regularization is employed to reinforce the discriminability of graph and avoid the trivial solution. Last but not least, an efficient iterative optimization algorithm is provided to optimize proposed model with a NP-hard constraint. Extensive experiments conducted on synthetic and several real-world data sets demonstrate the efficiency in classification task and robustness to noise of proposed method. (C) 2020 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据