417,398 Hits in 3.4 sec

Global and Local Structure Preserving Sparse Subspace Learning: An Iterative Approach to Unsupervised Feature Selection [article]

Nan Zhou, Yangyang Xu, Hong Cheng, Jun Fang, Witold Pedrycz
2015 arXiv   pre-print
In this paper, we propose a global and local structure preserving sparse subspace learning (GLoSS) model for unsupervised feature selection.  ...  The model can simultaneously realize feature selection and subspace learning.  ...  The Proposed Framework of Local Structure Preserving Sparse Subspace Learning In this section, we introduce our feature selection models that encourage global data fitting and also preserve local structure  ... 
arXiv:1506.01060v2 fatcat:f7kew7i6zjglfax37jnyjbdbfi

Global and Local Structure Preservation for Feature Selection

2014 IEEE Transactions on Neural Networks and Learning Systems  
In order to show this, we propose a global and local structure preservation framework for feature selection (GLSPFS) which integrates both global pairwise sample similarity and local geometric data structure  ...  scenarios, preserving global pairwise similarity is more important than preserving local geometric data structure; (3) In the unsupervised scenario, preserving local geometric data structure becomes clearly  ...  The typical algorithms include similarity preserving feature selection (SPFS) [17] , local learning based feature selection (LLFS) [18] , minimum redundancy maximum relevance (mRMR) [3] , robust feature  ... 
doi:10.1109/tnnls.2013.2287275 fatcat:dquf723ip5fw3pcfevjlfsuucq

Global and local structure preserving sparse subspace learning: An iterative approach to unsupervised feature selection

Nan Zhou, Yangyang Xu, Hong Cheng, Jun Fang, Witold Pedrycz
2016 Pattern Recognition  
In this paper, we propose a global and local structure preserving sparse subspace learning (GLoSS) model for unsupervised feature selection.  ...  The model can simultaneously realize feature selection and subspace learning.  ...  Sect. 3 reviews two local structure preserving methods and proposes a local structure preserving sparse subspace learning model.  ... 
doi:10.1016/j.patcog.2015.12.008 fatcat:hdjih2nkfnfghobxjy74pfuagy

Cross-View Local Structure Preserved Diversity and Consensus Learning for Multi-View Unsupervised Feature Selection

Chang Tang, Xinzhong Zhu, Xinwang Liu, Lizhe Wang
In this paper, we propose a cross-view local structure preserved diversity and consensus semantic learning model for MV-UFS, termed CRV-DCL briefly, to address these issues.  ...  learning tasks.  ...  Conclusions This paper introduces a novel MV-UFS method via crossview local structure preserved diversity and consensus learning.  ... 
doi:10.1609/aaai.v33i01.33015101 fatcat:vsua752cmnh7lhm6rwypfihdeu

Salient coding for image classification

Yongzhen Huang, Kaiqi Huang, Yinan Yu, Tieniu Tan
2011 CVPR 2011  
 Densely sampled SIFT features, dictionary size :2048, spatial pyramid feature, max-pooling.  Lib-linear SVM, average precision (AP)  ...  SALIENCY AWARE LOCALITY-PRESERVING CODING FOR IMAGE CLASSIFICATION Key Reference Locality-preserving Dictionary Learning  To learn a locality-preserving dictionary regarding the saliency characteristic  ...  .  Approach  A locality-preserving learning scheme is proposed by exploring the local geometrical structure around dictionary atom for dictionary creation.  Local density distribution is considered  ... 
doi:10.1109/cvpr.2011.5995682 dblp:conf/cvpr/HuangHYT11 fatcat:hmdkfcli7vcyzn6bamxnhvdl3i

Generative approach to unsupervised deep local learning

Changlu Chen, Chaoxi Niu, Xia Zhan, Kun Zhan
2019 Journal of Electronic Imaging (JEI)  
The locality-preserving loss designed by the constructed affinity graph serves as prior to preserve the local structure during the fine-tuning stage, which in turn improves the quality of feature representation  ...  Most existing feature learning methods optimize inflexible handcrafted features and the affinity matrix is constructed by shallow linear embedding methods.  ...  loss L locality for preserving local structure from pretrained GNN feature space to finetuned feature space.  ... 
doi:10.1117/1.jei.28.4.043005 fatcat:uq2cafqqpjbmdgbtghddgtuxee

Deep Clustering with Convolutional Autoencoders [chapter]

Xifeng Guo, Xinwang Liu, En Zhu, Jianping Yin
2017 Lecture Notes in Computer Science  
the local structure of data generating distribution in the learned feature space.  ...  Experiments on benchmark datasets empirically validate the power of convolutional autoencoders for feature learning and the effectiveness of local structure preservation.  ...  Our key idea is that CAE is beneficial to learning features for images and preserving local structure of data avoids distortion of feature space.  ... 
doi:10.1007/978-3-319-70096-0_39 fatcat:j2365mmpona4fchnxvvo3saoly

Locality-preserving K-SVD Based Joint Dictionary and Classifier Learning for Object Recognition

Yuan-Shan Lee, Chien-Yao Wang, Seksan Mathulaprangsan, Jia-Hao Zhao, Jia-Ching Wang
2016 Proceedings of the 2016 ACM on Multimedia Conference - MM '16  
For testing, additional information about the locality of query samples is obtained by treating the locality-preserving matrix as a feature.  ...  In the learning phase, the discriminative locality-preserving K-SVD (DLP-KSVD) in which the label information is incorporated into the locality-preserving term is proposed.  ...  To preserve the local structure of data in dictionary learning, Wei et al. [15] proposed a locality-sensitive SRC (L-SRC), which has a close-form solution throughout the learning process. Liu et al.  ... 
doi:10.1145/2964284.2967267 dblp:conf/mm/LeeWMZW16 fatcat:rx7htjr4afgcbaqrixvln2ecue

Locality Preserved Joint Nonnegative Matrix Factorization for Speech Emotion Recognition

2019 IEICE transactions on information and systems  
This study presents a joint dictionary learning approach for speech emotion recognition named locality preserved joint nonnegative matrix factorization (LP-JNMF).  ...  The learned representations are shared between the learned dictionaries and annotation matrix. Moreover, a locality penalty term is incorporated into the objective function.  ...  The locality-learned bases P map the activation matrix H into locality preserving code space. W is the linear classifier.  ... 
doi:10.1587/transinf.2018dal0002 fatcat:3atsfbqtenaunh5tawcr4sdx74

Saliency Aware Locality-preserving Coding for Image Classification

Quan Fang, Jitao Sang, Changsheng Xu
2012 2012 IEEE International Conference on Multimedia and Expo  
In this paper, we propose a saliency aware locality-preserving coding scheme by explicitly considering saliency into the dictionary creation and feature coding stages.  ...  However, recent locality-preserving coding schemes do not account for the saliency characteristic during the process of generating the raw image representations.  ...  Through this procedure, we can learn a locality-preserving dictionary for feature coding in the next section. The algorithm of dictionary learning is summarized in Algorithm 1.  ... 
doi:10.1109/icme.2012.164 dblp:conf/icmcs/FangSX12 fatcat:ajfob5e5yrhf7k7ckr2padhjjq


Yinghao Deng, Anhui University of Science & Technology, Qianjin Zhao, Shuzhi Su, Ruonan Zhang, Penglian Gao, Anhui University of Science & Technology, Anhui University of Science & Technology, Anhui University of Science & Technology, Anhui University of Science & Technology
2020 International Journal of Applied Mathematics and Machine Learning  
Canonical Correlation Analysis (CCA) is a classical feature learning method, which is widely used in image recognition, information fusion, and affective computing and so on.  ...  In view of this issue, locality preserving canonical correlation analysis (LPCCA) is proposed, which overcomes the preservation of local geometrical structure in CCA.  ...  Principal component analysis (PCA) [5] and locality preserving projection (LPP) [6] are traditional feature learning methods based on single-modal data.  ... 
doi:10.18642/ijamml_7100122110 fatcat:3phvfxhzvndjpifn2jdkvnoxum

Learning ordinal discriminative features for age estimation

Changsheng Li, Qingshan Liu, Jing Liu, Hanqing Lu
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
In this paper, we present a new method for facial age estimation based on ordinal discriminative feature learning.  ...  Considering the temporally ordinal and continuous characteristic of aging process, the proposed method not only aims at preserving the local manifold structure of facial images, but also it wants to keep  ...  The Importance of Preserving Locality Manifold learning uncovers the nonlinear structure by integrating the descriptions of a set of local patches using the neighborhood graph [16] .  ... 
doi:10.1109/cvpr.2012.6247975 dblp:conf/cvpr/LiLLL12 fatcat:f6je7wpo7fegtj75htsskm6mbi

Locality and similarity preserving embedding for feature selection

Xiaozhao Fang, Yong Xu, Xuelong Li, Zizhu Fan, Hong Liu, Yan Chen
2014 Neurocomputing  
., locality and similarity preserving embedding (LSPE) for feature selections.  ...  We impose ℓ2,1-norm on the transformation matrix to achieve row-sparsity, which allows us to select relevant features and learn the embedding simultaneously.  ...  Although the locality pays an important role in developing various kinds of algorithms, e.g., DR, semi-supervised learning algorithm, the features selected by the locality preserving-based feature selection  ... 
doi:10.1016/j.neucom.2013.08.040 fatcat:ujgaj3dcqvf4bgjtbaq2f6wgry

Local Deep-Feature Alignment for Unsupervised Dimension Reduction [article]

Jian Zhang, Jun Yu, Dacheng Tao
2019 arXiv   pre-print
This paper presents an unsupervised deep-learning framework named Local Deep-Feature Alignment (LDFA) for dimension reduction.  ...  We construct neighbourhood for each data sample and learn a local Stacked Contractive Auto-encoder (SCAE) from the neighbourhood to extract the local deep features.  ...  Therefore it could be important for the AEs to preserve the local characteristic during feature learning.  ... 
arXiv:1904.09747v1 fatcat:cjj2aq7efvhz7n3f7r5htnuxna

Improved Deep Embedded Clustering with Local Structure Preservation

Xifeng Guo, Long Gao, Xinwang Liu, Jianping Yin
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
By integrating the clustering loss and autoencoder's reconstruction loss, IDEC can jointly optimize cluster labels assignment and learn features that are suitable for clustering with local structure preservation  ...  Deep clustering learns deep feature representations that favor clustering task using neural networks.  ...  The autoencoder is used to learn representations in unsupervised manner and the learned features can preserve intrinsic local structure in data.  ... 
doi:10.24963/ijcai.2017/243 dblp:conf/ijcai/GuoGLY17 fatcat:fyprnqbqpfemzacs7ynxdfdbga
« Previous Showing results 1 — 15 out of 417,398 results