Filters








9,452 Hits in 6.3 sec

Factorized Graph Representations for Semi-Supervised Learning from Sparse Data [article]

Krishna Kumar P. and Paul Langton and Wolfgang Gatterbauer
2020 pre-print
Our approach first creates multiple factorized graph representations (with size independent of the graph) and then performs estimation on these smaller graph sketches.  ...  Can we instead directly estimate the correct compatibilities from a sparsely labeled graph in a principled and scalable way?  ...  Semi-Supervised Learning (SSL) Traditional graph-based Semi-Supervised Learning (SSL) predict the labels of unlabeled nodes under the assumption of homophily or smoothness.  ... 
doi:10.1145/3318464.3380577 arXiv:2003.02829v1 fatcat:jmfc4lvkrzd6jh5fl644jy6fda

Semi-supervised Data Representation via Affinity Graph Learning [article]

Weiya Ren
2015 arXiv   pre-print
A new semi-supervised learning framework is proposed by combing manifold regularization and data representation methods such as Non negative matrix factorization and sparse coding.  ...  The proposed framework forms the Laplacian regularizer through learning the affinity graph.  ...  The Euclidian distance based GNMF algorithms is: 2) Semi-supervised Graph Regularized Sparse Coding Graph regularized Sparse Coding (GSC) learns the sparse representations that explicitly take into account  ... 
arXiv:1502.03879v1 fatcat:badyxyo4pbf27mgyvt7tsyuqha

Feature Interaction-aware Graph Neural Networks [article]

Kaize Ding, Yichuan Li, Jundong Li, Chenghao Liu, Huan Liu
2020 arXiv   pre-print
However, most real-world graphs often come with high-dimensional and sparse node features, rendering the learned node representations from existing GNN architectures less expressive.  ...  In this paper, we propose Feature Interaction-aware Graph Neural Networks (FI-GNNs), a plug-and-play GNN framework for learning node representations encoded with informative feature interactions.  ...  For instance, the cross-entropy over all labeled data is employed as the loss function of the semi-supervised node classification problem [Kipf and Welling, 2016] : L semi = − 1 C l∈Y L C c=1 Y lc log  ... 
arXiv:1908.07110v2 fatcat:fk76u5vxorfljom5s3ivwhk6ku

Image Representation Learning Using Graph Regularized Auto-Encoders [article]

Yiyi Liao, Yue Wang, Yong Liu
2014 arXiv   pre-print
We consider the problem of image representation for the tasks of unsupervised learning and semi-supervised learning.  ...  To overcome this problem, the raw image vectors should be mapped to a proper representation space which can capture the latent structure of the original data and represent the data explicitly for further  ...  Table 6 : Results of the semi-supervised learning tasks for ORL. Experiments in Semi-supervised Learning For semi-supervised learning task, a small part of the samples are labeled.  ... 
arXiv:1312.0786v2 fatcat:4bopdzckm5eddiqcrltj4y3yde

A Survey on Concept Factorization: From Shallow to Deep Representation Learning [article]

Zhao Zhang, Yan Zhang, Mingliang Xu, Li Zhang, Yi Yang, Shuicheng Yan
2021 arXiv   pre-print
As a relatively new paradigm for representation learning, Concept Factorization (CF) has attracted a great deal of interests in the areas of machine learning and data mining for over a decade.  ...  Finally, we point out some future directions for studying the CF-based representation learning.  ...  ACKNOWLEDGMENT This work is partially supported by the National Natural Science Foundation of China (61672365) and the Fundamental Research Funds for the Central Universities of China (JZ2019H-  ... 
arXiv:2007.15840v3 fatcat:ahun2mogmfapxe4mqhqlsakyku

Image annotation bykNN-sparse graph-based label propagation over noisily tagged web images

Jinhui Tang, Richang Hong, Shuicheng Yan, Tat-Seng Chua, Guo-Jun Qi, Ramesh Jain
2011 ACM Transactions on Intelligent Systems and Technology  
To annotate the images more accurately, we propose a novel k NN-sparse graph-based semi-supervised learning approach for harnessing the labeled and unlabeled data simultaneously.  ...  The sparse graph constructed by datum-wise one-vs-k NN sparse reconstructions of all samples can remove most of the semantically-unrelated links among the data, and thus it is more robust and discriminative  ...  Label Reconstruction [Wright et al. 2009] SGSSL Sparse Graph-based Semi-Supervised Learning [Tang et al. 2009] k NN-SGSSL k NN-Sparse Graph-based Semi-Supervised Learning Ak NN-SGSSL Approximate  ... 
doi:10.1145/1899412.1899418 fatcat:tcwxxjmycjguzhvtuqocwbfqje

Combining graph embedding and sparse regression with structure low-rank representation for semi-supervised learning

Cong-Zhe You, Vasile Palade, Xiao-Jun Wu
2016 Complex Adaptive Systems Modeling  
Abstract In this paper, we propose a novel method for semi-supervised learning by combining graph embedding and sparse regression, termed as graph embedding and sparse regression with structure low rank  ...  Most of the graph based semi-supervised learning methods take into account the local neighborhood information while ignoring the global structure of the data.  ...  Acknowledgements The authors would like to thank the anonymous reviewers and editors for their valuable suggestions. Competing interests The authors declare that they have no competing interests.  ... 
doi:10.1186/s40294-016-0034-7 fatcat:lfspsce7bfeapfffawobaq7kwy

Disjoint Label Space Transfer Learning with Common Factorised Space [article]

Xiaobin Chang, Yongxin Yang, Tao Xiang, Timothy M. Hospedales
2018 arXiv   pre-print
It is particularly effective in handling a challenging case, where source and target label-spaces are disjoint, and outperforms alternatives in both unsupervised and semi-supervised settings.  ...  The key ingredient is a common representation termed Common Factorised Space.  ...  Semi-supervised Learning Graph-based regularisation is popular for semi-supervised learning (SSL) which uses both labelled and unlabelled data to achieve better performance than learning with labelled  ... 
arXiv:1812.02605v1 fatcat:o26mlpwkbfcxbh5nirdy3fhnt4

Disjoint Label Space Transfer Learning with Common Factorised Space

Xiaobin Chang, Yongxin Yang, Tao Xiang, Timothy M. Hospedales
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
It is particularly effective in handling a challenging case, where source and target label-spaces are disjoint, and outperforms alternatives in both unsupervised and semi-supervised settings.  ...  The key ingredient is a common representation termed Common Factorised Space.  ...  Semi-supervised Learning Graph-based regularisation is popular for semi-supervised learning (SSL) which uses both labelled and unlabelled data to achieve better performance than learning with labelled  ... 
doi:10.1609/aaai.v33i01.33013288 fatcat:jjnhyseegzginhjhkqsrpykohe

Semi-Supervised Boosting using Similarity Learning Based on Modular Sparse Representation with Marginal Representation Learning of Graph Structure Self-Adaptive

Shuhua Xu, Fei Gao
2020 IEEE Access  
[13] utilized sparse representation to learn similarity, and proposed similarity learning based on sparse representation for semi-supervised boosting.  ...  prediction learning, sparse representation with adaptive graph structure learning, and the improved model of sparse weight calculation of related modules to improve similarity learning model of semi-supervised  ... 
doi:10.1109/access.2020.3030163 fatcat:xluipjbv65fbrphnlpchyvnigq

Robust Spectral Clustering via Sparse Representation [chapter]

Xiaodong Feng
2018 Recent Applications in Data Clustering  
Spectral clustering via sparse representation has been proposed for clustering high-dimensional data.  ...  Clustering high-dimensional data has been a challenging problem in data mining and machining learning.  ...  In semi-supervised learning by sparse representation [18] , the graph adjacency structure as well as the graph weights of the directed graph construction is derived simultaneously and in a parameter-free  ... 
doi:10.5772/intechopen.76586 fatcat:zcrqwtf6yba5niv2o74j6vvjnu

Structure Preserving Low-Rank Representation for Semi-supervised Face Recognition [chapter]

Yong Peng, Suhang Wang, Shen Wang, Bao-Liang Lu
2013 Lecture Notes in Computer Science  
shown excellent performance in semi-supervised learning.  ...  Constructing an informative and discriminative graph plays an important role in the graph based semi-supervised learning methods.  ...  for semi-supervised learning.  ... 
doi:10.1007/978-3-642-42042-9_19 fatcat:zfp3k66a2rge7dpcri5yyzb6my

Enhanced low-rank representation via sparse manifold adaption for semi-supervised learning

Yong Peng, Bao-Liang Lu, Suhang Wang
2015 Neural Networks  
Among the existing graph-based learning models, low-rank representation (LRR) is a very competitive one, which has been extensively employed in spectral clustering and semi-supervised learning (SSL).  ...  In this paper, we propose an enhanced LRR via sparse manifold adaption, termed manifold low-rank representation (MLRR), to learn low-rank data representation.  ...  Experimental results on four popular data sets showed that MLRR is a competitive model for graph-based semi-supervised learning.  ... 
doi:10.1016/j.neunet.2015.01.001 pmid:25634552 fatcat:xmy7g3owfza6vnw63vafnbomvu

Deep graph learning for semi-supervised classification [article]

Guangfeng Lin, Xiaobing Kang, Kaiyang Liao, Fan Zhao, Yajun Chen
2020 arXiv   pre-print
To simulate the interdependence, deep graph learning(DGL) is proposed to find the better graph representation for semi-supervised classification.  ...  GCN for semi-supervised classification.  ...  The authors would like to thank the anonymous reviewers for their insightful comments that help improve the quality of this paper. This work was supported by NSFC  ... 
arXiv:2005.14403v1 fatcat:oioedbp6cjg4bfu7chmf4cspqy

Should we discard sparse or incomplete videos?

Chuan Sun, Hassan Foroosh
2014 2014 IEEE International Conference on Image Processing (ICIP)  
unknown actions by a graph based semi-supervised learning framework.  ...  data as a semi-supervised learning problem of labeled and unlabeled data. (2) We introduce a two-step approach to convert the input mixed data into a uniform compact representation. (3) Exhaustively scrutinizing  ...  Thirdly, we build lower dimensional representation using a rank one tensor decomposition algorithm. Finally, we apply graph based semi-supervised learning for action classification.  ... 
doi:10.1109/icip.2014.7025506 dblp:conf/icip/SunF14 fatcat:5pxbrtyisrbn3ksbqvdfsv5anq
« Previous Showing results 1 — 15 out of 9,452 results