Filters








5,195 Hits in 3.2 sec

A Flexible Generative Framework for Graph-based Semi-supervised Learning [article]

Jiaqi Ma, Weijing Tang, Ji Zhu, Qiaozhu Mei
2019 arXiv   pre-print
In this work, we propose a flexible generative framework for graph-based semi-supervised learning, which approaches the joint distribution of the node features, labels, and the graph structure.  ...  We consider a family of problems that are concerned about making predictions for the majority of unlabeled, graph-structured data samples based on a small proportion of labeled samples.  ...  Graph-based Regularization for Semi-supervised Learning One of the most popular types of graph-based semi-supervised learning methods is the graph-based regularization methods.  ... 
arXiv:1905.10769v2 fatcat:mtooywzivfevricftsdcjr56ke

KGNN: Harnessing Kernel-based Networks for Semi-supervised Graph Classification [article]

Wei Ju, Junwei Yang, Meng Qu, Weiping Song, Jianhao Shen, Ming Zhang
2022 arXiv   pre-print
kernels to explicitly compare each query graph with all the labeled graphs stored in a memory for prediction.  ...  This paper studies semi-supervised graph classification, which is an important problem with various applications in social network analysis and bioinformatics.  ...  For graph-structured data, one fundamental problem is graph classification, which aims at analyzing and predicting the property of the entire graph.  ... 
arXiv:2205.10550v1 fatcat:shd52l46wbgsvptdtz3zqriy5a

Semi-supervised Max-margin Topic Model with Manifold Posterior Regularization

Wenbo Hu, Jun Zhu, Hang Su, Jingwei Zhuo, Bo Zhang
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
In this paper, we present an effective semi-supervised max-margin topic model by naturally introducing manifold posterior regularization to a regularized Bayesian topic model, named LapMedLDA.  ...  As collecting a fully labeled dataset is often time-consuming, semi-supervised learning is of high interest.  ...  Our Proposal In this paper, we propose a semi-supervised topic model with manifold posterior regularization.  ... 
doi:10.24963/ijcai.2017/259 dblp:conf/ijcai/HuZSZZ17 fatcat:xs4t3kasvbdm3ax4jpqips26ha

High Order Regularization for Semi-Supervised Learning of Structured Output Problems

Yujia Li, Richard S. Zemel
2014 International Conference on Machine Learning  
We propose a new max-margin framework for semi-supervised structured output learning, that allows the use of powerful discrete optimization algorithms and high order regularizers defined directly on model  ...  predictions for the unlabeled examples.  ...  Acknowledgments We thank Charlie Tang and Danny Tarlow for helpful discussions.  ... 
dblp:conf/icml/LiZ14 fatcat:huvm4ronxjb3tiorhhwf746dta

Uncertainty Aware Graph Gaussian Process for Semi-Supervised Learning

Zhao-Yang Liu, Shao-Yuan Li, Songcan Chen, Yao Hu, Sheng-Jun Huang
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Graph-based semi-supervised learning (GSSL) studies the problem where in addition to a set of data points with few available labels, there also exists a graph structure that describes the underlying relationship  ...  UaGGP exploits the prediction uncertainty and label smooth regularization to guide each other during learning.  ...  Conclusion In this paper, we propose a Gaussian Process based approach UaGGP for semi-supervised classification with graph structures.  ... 
doi:10.1609/aaai.v34i04.5934 fatcat:jqxfb3isc5fuffivgclx7z4kea

Neighborhood Random Walk Graph Sampling for Regularized Bayesian Graph Convolutional Neural Networks [article]

Aneesh Komanduri, Justin Zhan
2021 arXiv   pre-print
Introducing the Bayesian paradigm to graph-based models, specifically for semi-supervised node classification, has been shown to yield higher classification accuracies.  ...  algorithm utilizing graph structure, reduces overfitting by using a variational inference layer, and yields consistently competitive classification results compared to the state-of-the-art in semi-supervised  ...  based on graph structure.  ... 
arXiv:2112.07743v1 fatcat:jy5jjzyywjfcvjpqmhkwhxls2i

Semi-supervised Deep Kernel Learning: Regression with Unlabeled Data by Minimizing Predictive Variance [article]

Neal Jean, Sang Michael Xie, Stefano Ermon
2019 arXiv   pre-print
We present semi-supervised deep kernel learning (SSDKL), a semi-supervised regression model based on minimizing predictive variance in the posterior regularization framework.  ...  for regression.  ...  We are thankful to Volodymyr Kuleshov and Aditya Grover for helpful discussions.  ... 
arXiv:1805.10407v4 fatcat:47lcumjrhnaidj3c5nlceb72ty

Efficient Graph-Based Semi-Supervised Learning of Structured Tagging Models

Amarnag Subramanya, Slav Petrov, Fernando C. N. Pereira
2010 Conference on Empirical Methods in Natural Language Processing  
We describe a new scalable algorithm for semi-supervised training of conditional random fields (CRF) and its application to partof-speech (POS) tagging.  ...  The similarity graph is used during training to smooth the state posteriors on the target domain. Standard inference can be used at test time.  ...  Altun et al. (2005) proposed a max-margin objective for semi-supervised learning over structured spaces.  ... 
dblp:conf/emnlp/SubramanyaPP10 fatcat:imoqrbvfv5bk3nstevm2wvop4e

Domain Constraint Approximation based Semi Supervision [article]

Yifu Wu, Jin Wei, Rigoberto Roche
2019 arXiv   pre-print
In this paper, we proposed a fuzzy domain-constraint-based framework which loses the requirement of traditional constraint learning and enhances the model quality for semi supervision.  ...  Domain constraint is another way regularize the posterior but has some limitation.  ...  Conclusion In this paper, a fuzzy domain-knowledge-based framework of posterior regularization is proposed for semi supervision.  ... 
arXiv:1902.04177v2 fatcat:wjbcqspbq5ac7ainswqogvrjma

A Survey on Deep Semi-supervised Learning [article]

Xiangli Yang, Zixing Song, Irwin King, Zenglin Xu
2021 arXiv   pre-print
We first present a taxonomy for deep semi-supervised learning that categorizes existing methods, including deep generative methods, consistency regularization methods, graph-based methods, pseudo-labeling  ...  Deep semi-supervised learning is a fast-growing field with a range of practical applications.  ...  Based on graph-based semi-supervised methods, [257] copes with various degrees of class imbalance in a given dataset. Robust semi-supervised learning.  ... 
arXiv:2103.00550v2 fatcat:lymncf5wavgkhaenbvqlyvhuaa

Bayesian Semi-supervised Learning with Graph Gaussian Processes [article]

Yin Cheng Ng, Nicolo Colombo, Ricardo Silva
2018 arXiv   pre-print
We propose a data-efficient Gaussian process-based Bayesian approach to the semi-supervised learning problem on graphs.  ...  The proposed model shows extremely competitive performance when compared to the state-of-the-art graph neural networks on semi-supervised learning benchmark experiments, and outperforms the neural networks  ...  Many of the successful graph-based semi-supervised learning models are based on graph Laplacian regularization or learning embeddings of the nodes.  ... 
arXiv:1809.04379v3 fatcat:ynsxjoq4k5gdbk7ziesvm77leu

Wikipedia entity expansion and attribute extraction from the web using semi-supervised learning

Lidong Bing, Wai Lam, Tak-Lam Wong
2013 Proceedings of the sixth ACM international conference on Web search and data mining - WSDM '13  
Then the semi-supervised learning process can leverage the unlabeled data in the record set by controlling the label regularization under the guidance of the proximate record graph.  ...  We make use of a proximate record graph to guide the semi-supervised learning process. The graph captures alignment similarity among data records.  ...  Posterior Regularization In Lines 7 and 8 of the semi-supervised learning model shown in Figure 5 , the posterior label distribution is regularized with the guidance of the proximate graph G.  ... 
doi:10.1145/2433396.2433468 dblp:conf/wsdm/BingLW13 fatcat:wscamx4trbftdh6kdonbyrczlu

Virtual Adversarial Training on Graph Convolutional Networks in Node Classification [article]

Ke Sun, Zhouchen Lin, Hantao Guo, Zhanxing Zhu
2020 arXiv   pre-print
By imposing virtually adversarial smoothness on the posterior distribution in semi-supervised learning, VAT yields improvement on the Symmetrical Laplacian Smoothness of GCNs.  ...  The effectiveness of Graph Convolutional Networks (GCNs) has been demonstrated in a wide range of graph-based machine learning tasks.  ...  Effect of Regularization in Semi-Supervised Learning. Regularization plays a crucial role in semi-supervised learning including graph-based learning tasks.  ... 
arXiv:1902.11045v2 fatcat:gk5bqvky5rcjlicsuyuycunehe

Graph Based Multi-class Semi-supervised Learning Using Gaussian Process [chapter]

Yangqiu Song, Changshui Zhang, Jianguo Lee
2006 Lecture Notes in Computer Science  
This paper proposes a multi-class semi-supervised learning algorithm of the graph based method. We make use of the Bayesian framework of Gaussian process to solve this problem.  ...  We propose the prior based on the normalized graph Laplacian, and introduce a new likelihood based on softmax function model.  ...  If the data present explicit structure of a manifold, the graph based semi-supervised learning algorithms work efficiently for both transducitve and inductive problems.  ... 
doi:10.1007/11815921_49 fatcat:6i45ryfkvzfjjgyjta57a2il6m

Entropic Graph-based Posterior Regularization

Maxwell W. Libbrecht, Michael M. Hoffman, Jeff A. Bilmes, William Stafford Noble
2015 International Conference on Machine Learning  
We define a new class of entropic graph-based posterior regularizers that augment a probabilistic model by encouraging pairs of nearby variables in a regularization graph to have similar posterior distributions  ...  Graph smoothness objectives have achieved great success in semi-supervised learning but have not yet been applied extensively to unsupervised generative models.  ...  For SQGPR and EGPR, we used a GBR graph based on 3D structure data.  ... 
dblp:conf/icml/LibbrechtHBN15 fatcat:cmwypsm4cnc3fcba5zlwrup7cq
« Previous Showing results 1 — 15 out of 5,195 results