Positive and Negative Label Propagations

Zoidi, Olga; Tefas, Anastasios; Nikolaidis, Nikos; Pitas, Ioannis

doi:10.1109/tcsvt.2016.2598671

Cited by 25 publications

(17 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As is common practice, each face/object image is resized to 32×32 pixels; hence, each image corresponds to a data point in a 1024-dimensional space. For the transductive classification on each database, each database is split into a labeled set and an unlabeled set, similar to [1][2][3][4][5][6][7][8][9] [21]. Finally, the accuracy is computed by comparing the predicted labels of the unlabeled samples with the ground-truth labels that are provided by the original data corpus.…”

Section: Simulation Results and Analysismentioning

confidence: 99%

“…is the trace operator, T is the transpose of the corresponding matrix, and U denotes a diagonal matrix for balancing the similarity-based manifold smoothness and label fitness. The label of each sample i x is received partially from its neighbors and partially from the initial state [1][2][3][4][5][6][7][8][9]. Matrix U includes the weighting factors for the labeled and unlabeled data, which is defined as…”

Section: A Regular Transductive Label Propagationmentioning

confidence: 99%

“…Representative transductive LP algorithms mainly include Gaussian fields and harmonic function (GFHF) [1], linear neighborhood propagation (LNP) [2], special LP (SLP) [3], learning with local and global Consistency (LLGC) [5], projective label propagation (ProjLP) via label embedding [6], prior-class-dissimilarity-based LNP (CD-LNP) [4], sparse neighborhood propagation (SparseNP) [8], positive and negative label propagation (PN-LP) [7], and adaptive neighborhood propagation (AdaptiveNP) [9]. GFHF, LLGC, LNP and SLP optimize similar objective functions to predict the labels of samples by receiving information partly from the neighborhood and partly from the initial state.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Robust Triple-Matrix-Recovery-Based Auto-Weighted Label Propagation for Classification

Zhang

Zhao

et al. 2020

IEEE Trans. Neural Netw. Learning Syst.

View full text Add to dashboard Cite

The graph-based semi-supervised label propagation algorithm has delivered impressive classification results. However, the estimated soft labels typically contain mixed signs and noise, which cause inaccurate predictions due to the lack of suitable constraints. Moreover, available methods typically calculate the weights and estimate the labels in the original input space, which typically contains noise and corruption. Thus, the encoded similarities and manifold smoothness may be inaccurate for label estimation. In this paper, we present effective schemes for resolving these issues and propose a novel and robust semi-supervised classification algorithm, namely, the triple-matrix-recovery-based robust auto-weighted label propagation framework (ALP-TMR). Our ALP-TMR introduces a triple matrix recovery mechanism to remove noise or mixed signs from the estimated soft labels and improve the robustness to noise and outliers in the steps of assigning weights and predicting the labels simultaneously. Our method can jointly recover the underlying clean data, clean labels and clean weighting spaces by decomposing the original data, predicted soft labels or weights into a clean part plus an error part by fitting noise. In addition, ALP-TMR integrates the auto-weighting process by minimizing reconstruction errors over the recovered clean data and clean soft labels, which can encode the weights more accurately to improve both data representation and classification. By classifying samples in the recovered clean label and weight spaces, one can potentially improve the label prediction results. The results of extensive experiments demonstrated the satisfactory performance of our ALP-TMR.

show abstract

Section: Simulation Results and Analysismentioning

confidence: 99%

Section: A Regular Transductive Label Propagationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Robust Triple-Matrix-Recovery-Based Auto-Weighted Label Propagation for Classification

Zhang

Zhao

et al. 2020

IEEE Trans. Neural Netw. Learning Syst.

View full text Add to dashboard Cite

show abstract

“…PN-LP extends the existing LP framework to the scenario of label propagation with both positive and negative labels [13]. Given a set of samples…”

Section: A Positive and Negative Label Propagation (Pn-lp)mentioning

confidence: 99%

“…On one hand, for positive label propagation,  i u is usually set to a large value (e.g., +∞) for labeled data so that the resulted label can keep consistent with the initial states, and is set to be a small value (e.g., 0) for the unlabeled data so that label information from the neighbors can be received for class assignments [5]. On the other hand, for positive and negative label propagation, the significance of positive and negative labels is usually unequal, and the increased accuracy can be obtained when the significance of positive labels is higher than negative ones [13]. Based on the above analysis, in this work we set 10 10   i u (to approximate +∞) and 0 for labeled and unlabeled data respectively, and set 1   i u and 0 for the labeled and unlabeled data respectively in simulations.…”

Section: B Proposed Formulationmentioning

confidence: 99%

Kernel-Induced Label Propagation by Mapping for Semi-Supervised Classification

Zhang

Jia

Zhao

et al. 2019

IEEE Trans. Big Data

View full text Add to dashboard Cite

Kernel methods have been successfully applied to the areas of pattern recognition and data mining. In this paper, we mainly discuss the issue of propagating labels in kernel space. A Kernel-Induced Label Propagation (Kernel-LP) framework by mapping is proposed for high-dimensional data classification using the most informative patterns of data in kernel space. The essence of Kernel-LP is to perform joint label propagation and adaptive weight learning in a transformed kernel space. That is, our Kernel-LP changes the task of label propagation from the commonly-used Euclidean space in most existing work to kernel space. The motivation of our Kernel-LP to propagate labels and learn the adaptive weights jointly by the assumption of an inner product space of inputs, i.e., the original linearly inseparable inputs may be mapped to be separable in kernel space. Kernel-LP is based on existing positive and negative LP model, i.e. the effects of negative label information are integrated to improve the label prediction power. Also, Kernel-LP performs adaptive weight construction over the same kernel space, so it can avoid the tricky process of choosing the optimal neighborhood size suffered in traditional criteria. Two novel and efficient out-ofsample approaches for our Kernel-LP to involve new test data are also presented, i.e., (1) direct kernel mapping and (2) kernel mapping-induced label reconstruction, both of which purely depend on the kernel matrix between training set and testing set. Owing to the kernel trick, our algorithms will be applicable to handle the high-dimensional real data. Extensive results on real datasets demonstrate the effectiveness of our approach. Index Terms-Kernel-induced label propagation; mapping; adaptive weight learning; semi-supervised classification 1 I.Lei Jia is currently working toward the research degree at the School of Computer Science and Technology, Soochow University, China. His current research interests include machine learning, pattern recognition, data mining and their real applications. More specifically, he is interested in designing effective semi-supervised learning algorithms for classification. He has published several papers in Neural Networks, and conference proceedings of ICDM, ICONIP and PCM.Mingbo Zhao (M'13-) received the Ph.D. degree from the

show abstract

Robust and sparse label propagation for graph-based semi-supervised classification

Hua

Yang

2021

Appl Intell

View full text Add to dashboard Cite

Positive and Negative Label Propagations

Cited by 25 publications

References 39 publications

Robust Triple-Matrix-Recovery-Based Auto-Weighted Label Propagation for Classification

Robust Triple-Matrix-Recovery-Based Auto-Weighted Label Propagation for Classification

Kernel-Induced Label Propagation by Mapping for Semi-Supervised Classification

Robust and sparse label propagation for graph-based semi-supervised classification

Contact Info

Product

Resources

About