Semi-supervised Learning Method Research Articles

Handling the Challenges of Small-Scale Labeled Data and Class Imbalances in Classifying the N and K Statuses of Rubber Leaves Using Hyperspectroscopy Techniques.

The nutritional status of rubber trees (Hevea brasiliensis) is inseparable from the production of natural rubber. Nitrogen (N) and potassium (K) levels in rubber leaves are 2 crucial criteria that reflect the nutritional status of the rubber tree. Advanced hyperspectral technology can evaluate N and K statuses in leaves rapidly. However, high bias and uncertain results will be generated when using a small size and imbalance dataset to train a spectral estimaion model. A typical solution of laborious long-term nutrient stress and high-intensive data collection deviates from rapid and flexible advantages of hyperspectral tech. Therefore, a less intensive and streamlined method, remining information from hyperspectral image data, was assessed. From this new perspective, a semisupervised learning (SSL) method and resampling techniques were employed for generating pseudo-labeling data and class rebalancing. Subsequently, a 5-classification spectral model of the N and K statuses of rubber leaves was established. The SSL model based on random forest classifiers and mean sampling techniques yielded optimal classification results both on imbalance/balance dataset (weighted average precision 67.8/78.6%, macro averaged precision 61.2/74.4%, and weighted recall 65.7/78.5% for the N status). All data and code could be viewed on the:Github https://github.com/WeehowTang/SSL-rebalancingtest. Ultimately, we proposed an efficient way to rapidly and accurately monitor the N and K levels in rubber leaves, especially in the scenario of small annotation and imbalance categories ratios.

Enhancing collaborative learning in psychoeducation through social network analysis

Abstract Social network opens up a practical new path for psychoeducation and, with the help of a multimedia mobile terminal’s network, text sentiment analysis for students’ mental health problem revelation and constructing corresponding psychoeducation model. Based on the above viewpoint, this paper proposes a semi-supervised learning method for sentiment analysis using graphs. Specifically, a graph-based semi-supervised learning classifier is constructed using the “Weibo-Weibo” relationship graph. At the same time, social network and text similarity relationships are combined. A link between annotated and unannotated texts was established, and an optimization algorithm was used to solve the model, revealing the psychological problems of students under the emotional texts of social networks, thus constructing a corresponding “four-in-one” mental health education model, which enhances the collaborative learning of students and the four major organizations in psychoeducation. The accuracy, accuracy, recall, and F1 value are outperformed by the model in this paper by 2.44%, 2.11%, 1.45%, and 2.34%, respectively. The eight questions designed around the three dimensions of collaborative learning situation, learning effectiveness, and four-in-one teaching model were highly rated by students. The four-in-one teaching model that utilizes social network sentiment analysis effectively enhances students’ collaborative learning and improves their ability to solve psychoeducation’s difficulties.

Dual uncertainty-guided multi-model pseudo-label learning for semi-supervised medical image segmentation.

Semi-supervised medical image segmentation is currently a highly researched area. Pseudo-label learning is a traditional semi-supervised learning method aimed at acquiring additional knowledge by generating pseudo-labels for unlabeled data. However, this method relies on the quality of pseudo-labels and can lead to an unstable training process due to differences between samples. Additionally, directly generating pseudo-labels from the model itself accelerates noise accumulation, resulting in low-confidence pseudo-labels. To address these issues, we proposed a dual uncertainty-guided multi-model pseudo-label learning framework (DUMM) for semi-supervised medical image segmentation. The framework consisted of two main parts: The first part is a sample selection module based on sample-level uncertainty (SUS), intended to achieve a more stable and smooth training process. The second part is a multi-model pseudo-label generation module based on pixel-level uncertainty (PUM), intended to obtain high-quality pseudo-labels. We conducted a series of experiments on two public medical datasets, ACDC2017 and ISIC2018. Compared to the baseline, we improved the Dice scores by 6.5% and 4.0% over the two datasets, respectively. Furthermore, our results showed a clear advantage over the comparative methods. This validates the feasibility and applicability of our approach.

Supervision by Denoising.

Learning-based image reconstruction models, such as those based on the U-Net, require a large set of labeled images if good generalization is to be guaranteed. In some imaging domains, however, labeled data with pixel- or voxel-level label accuracy are scarce due to the cost of acquiring them. This problem is exacerbated further in domains like medical imaging, where there is no single ground truth label, resulting in large amounts of repeat variability in the labels. Therefore, training reconstruction networks to generalize better by learning from both labeled and unlabeled examples (called semi-supervised learning) is problem of practical and theoretical interest. However, traditional semi-supervised learning methods for image reconstruction often necessitate handcrafting a differentiable regularizer specific to some given imaging problem, which can be extremely time-consuming. In this work, we propose "supervision by denoising" (SUD), a framework to supervise reconstruction models using their own denoised output as labels. SUD unifies stochastic averaging and spatial denoising techniques under a spatio-temporal denoising framework and alternates denoising and model weight update steps in an optimization framework for semi-supervision. As example applications, we apply SUD to two problems from biomedical imaging-anatomical brain reconstruction (3D) and cortical parcellation (2D)-to demonstrate a significant improvement in reconstruction over supervised-only and ensembling baselines. Our code available at https://github.com/seannz/sud.

Deep Semi-Supervised Learning Method for False Data Detection Against Forgery and Concealing of Faults in Cyber-Physical Power Systems

This paper proposes a deep semi-supervised numerical false data detection (DSS-NFDD) method, where hidden data in labeled and unlabeled datasets are leveraged simultaneously, to detect the false data in CPPS. Two types of false data are considered in this study: one is the forged fault data, which makes operators mistakenly believe that there is a fault in their operating system; the other is the false data used to conceal actual faults. First, a data dimension reduction method is proposed based on the PageRank algorithm to avoid the excessive noise caused by high-dimension data. Then, a semi-supervised deep learning framework is established to detect false data samples, which consists of two parts: one is the priori estimation module, and the other is a false data scoring network. A novel concentration loss function is presented for training the false data scoring network, which minimizes the impacts of noise pollution and sample bias. Next, a false feature location method is proposed to help human operators eliminate anomalies. Finally, the effectiveness and the superiorities of the proposed DSS-NFDD method are verified by analyzing the simulation results for the IEEE-39 bus and IEEE-118 bus systems.

Порівняння ефективності методів напівкерованого навчання на основі автоенкодерів для задач класифікації фотографій ракових захворювань шкіри

Порівняння ефективності методів напівкерованого навчання на основі автоенкодерів для задач класифікації фотографій ракових захворювань шкіри

Handling Labeled Data Insufficiency: Semi-supervised Learning with Self-Training Mixup Decision Tree for Classification of Network Attacking Traffic

Motivated by the fast advancements in artificial intelligence (AI) technologies, recent research has moved towards using machine learning and deep learning to detect and classify security attacks in computer networks. However, most prior works adopt supervised learning methods, and the performance heavily depends on the amount of labeled data used to train the detection models. Network attack detection and classification is not an exception due to the lack of labeled data, especially the attacking traffic, which is much less than the regular (legitimate) traffic. Yet, labeling network traffic is also challenging and requires specific domain expertise. This paper proposes an efficient semi-supervised learning method for the classification of network attacking traffic, known as Self-Training Mixup Decision Tree (STM-DT). STM-DT first trains a decision tree on a small amount of labeled data and then uses the obtained model to predict labels of unlabeled samples. Some noisy labels will be removed by consistency. The predicted samples will then be mixed with labeled samples using <monospace>mixup</monospace> to train a new decision tree, which is the final desired classifier. We evaluate STM-DT using four network traffic datasets. Experimental results demonstrate that the proposed STM-DT method achieves higher macro F1 scores over different minority labeled data percentages.

SemiRS-COC: Semi-Supervised Classification for Complex Remote Sensing Scenes with Cross-Object Consistency.

Semi-supervised learning (SSL), which aims to learn with limited labeled data and massive amounts of unlabeled data, offers a promising approach to exploit the massive amounts of satellite Earth observation images. The fundamental concept underlying most state-of-the-art SSL methods involves generating pseudo-labels for unlabeled data based on image-level predictions. However, complex remote sensing (RS) scene images frequently encounter challenges, such as interference from multiple background objects and significant intra-class differences, resulting in unreliable pseudo-labels. In this paper, we propose the SemiRS-COC, a novel semi-supervised classification method for complex RS scenes. Inspired by the idea that neighboring objects in feature space should share consistent semantic labels, SemiRS-COC utilizes the similarity between foreground objects in RS images to generate reliable object-level pseudo-labels, effectively addressing the issues of multiple background objects and significant intra-class differences in complex RS images. Specifically, we first design a Local Self-Learning Object Perception (LSLOP) mechanism, which transforms multiple background objects interference of RS images into usable annotation information, enhancing the model's object perception capability. Furthermore, we present a Cross-Object Consistency Pseudo-Labeling (COCPL) strategy, which generates reliable object-level pseudo-labels by comparing the similarity of foreground objects across different RS images, effectively handling significant intra-class differences. Extensive experiments demonstrate that our proposed method achieves excellent performance compared to state-of-the-art methods on three widely-adopted RS datasets.

DSST: A dual student model guided student–teacher framework for semi-supervised medical image segmentation

Numerous semi-supervised learning methods are emerging in medical image segmentation to reduce the dependency of deep learning models on pixel-level annotation data. Consistency regularization methods based on the Student–Teacher structure have achieved brilliant results in this domain. However, the current structures are unable to resolve the tight weight coupling satisfactorily between the teacher and student model, which leads to a decrease in the segmentation performance. In this paper, we propose a novel and practical semi-supervised learning framework, Dual-Student-Single-Teacher (DSST), to alleviate this problem. Particularly, the DSST framework consists of three segmentation models with the identical structure but different initial parameters, one serves as the teacher model and others as the student models, which employs an alternating manner to update the teacher model parameters. For the DSST framework, we present different supervised modes to sufficiently explore the enhancement of consistency regularization for model segmentation performance. Furthermore, we also introduce abundant and efficient input and feature perturbations for the proposed method. Finally, we evaluate our framework in three public medical image segmentation tasks, including Pancreas-CT, LA dataset, and cardiac segmentation on the ACDC dataset. Extensive experiments demonstrate that, compared with eight other superior semi-supervised methods, the DSST method obtains state-of-the-art segmentation performance and is an effective and generalizable framework. Code is available at https://github.com/LBL0704/DSST.

DYMatch: Semi-Supervised Learning with Dynamic Pseudo Labeling and Feature Consistency

A considerable number of approaches based on consistency regularization and pseudo-labeling have been proposed in Semi-supervised Learning (SSL) so far. These approaches significantly enhance the SSL methods by effectively utilizing large amount of unlabeled data to improve the model’s performance. However, existing methods may fail to utilize the unlabeled data more efficiently, mainly due to the challenges faced in pseudo-label estimation. In this paper, we begin by analyzing the impact of the pseudo-label estimation method on training the SSL model. Furthermore, we emphasize that an effective pseudo-label estimation method should reflect the difference of recognition performance among samples from different categories, and also ensure the maintenance of both high-quantity and high-quality pseudo-labels during each training iteration. Based on above analysis, we propose DYMatch, an innovative SSL method employing a dynamic estimation process. Firstly, a dynamic pseudo-label estimation method based on Gaussian mixture model is proposed to dynamically estimate the confidence threshold on different categories. Secondly, a feature-correlation consistency regularization method is introduced to further enhance the learning on unlabeled data. The experimental results show that DYMatch is a simple and effective SSL method, especially when the labeled data is limited.

Discrimination of the degree of heavy metal pollution in corn leaves in mining areas based on semi-supervised learning

ABSTRACT To assess the degree of heavy metal pollution in crops, this study focused on crop leaves subjected to varying levels of heavy metal copper (Cu) stress. After removing outliers and applying smoothing techniques, the spectral data underwent derivative (D) and multiplicative scatter correction (MSC). Competitive adaptive reweighted sampling (CARS) and specific spectral ranges (SSR) were employed to extract the feature bands. Six different combinations of data preprocessing methods were utilized. This model incorporates the semi-supervised learning method, which enables the reduction of time and cost associated with annotating large-scale data. It addresses the identification problem when the labelled data is limited. Finally, the eXtreme Gradient Boosting algorithm (XGBoost) is used to select the best model and discriminate the degree of heavy metal pollution in corn. The D-CARS-XGBoost model achieved an accuracy rate exceeding 98%.

MLC: Multi-level consistency learning for semi-supervised left atrium segmentation

Atrial fibrillation is the most common type of arrhythmia associated with a high mortality rate. Left atrium segmentation is crucial for the diagnosis and treatment of atrial fibrillation. Accurate left atrium segmentation with limited labeled data is a tricky problem. In this paper, a novel multi-level consistency semi-supervised learning method is proposed for left atrium segmentation from 3D magnetic resonance images. The proposed framework can efficiently utilize limited labeled data and large amounts of unlabeled data by performing consistency predictions under task level, data level, and feature level perturbations. For task consistency, the segmentation results and signed distance maps were used for both segmentation and distance estimation tasks. For data level perturbation, random flips (horizontal or vertical) were introduced for unlabeled data. Moreover, based on virtual adversarial training, we design a multi-layer feature perturbation in the structure of skipping connection. Our method is evaluated on the publicly available Left Atrium Segmentation Challenge dataset version 2018. For the model trained with a label rate of 20%, the evaluation metrics Dice, Jaccard, ASD, and 95HD are 91.69%, 84.71%, 1.43 voxel, and 5.44 voxel, respectively. The experimental results show that the proposed method outperforms other semi-supervised learning methods and even achieves better performance than the fully supervised V-Net.

Analysis of SARS-CoV-2 transmission in airports based on real human close contact behaviors

The COVID-19 pandemic has significantly impacted people's daily lives for over three years. Airports, with their dense population and frequent close contact, pose a higher risk of respiratory infectious diseases compared to many other indoor environments. However, limited availability of data on close contact behavior has resulted in a gap in indoor exposure analysis. This study conducted depth sensor measurements and video data collection across nine areas of a northern (airport A) and a southern (airport B) airports in China by 11 participants. The data, comprising more than 44 h of close contact behaviors, including interpersonal distance, relative facial orientation, and the relative position of individuals, were analyzed using a semi-supervised machine learning method. Based on this analysis, a close contact transmission model for COVID-19 was developed, which considers the aforementioned close contact behaviors to assess the risk of exposure and the efficacy of interventions. The average close contact ratio in 9 airport's areas is 25.4 % (ranging from 6.1 % to 55.0 %), with passengers having the highest frequency of close contact in manual check-in areas. During close contacts, the average interpersonal distance in airports is 1.2 m (ranging from 1.1 to 1.4 m), being shortest in boarding areas. Face-to-face close contact is highest in charging areas, with a percentage of 46.9 %. If people maintain a distance of over 1.0 m in all areas, the total virus exposure could be reduced by 6.9 %–22.0 % compared to the actual situation. Dining areas have the highest virus exposure risk for both short-range inhalation and mucosal deposition, followed by manual check-in areas. This study provides a data support for the scientific epidemic prevention and control in airports from the viewpoint of close contact behaviors.

Semi-supervised liver segmentation based on local regions self-supervision.

Semi-supervised learning has gained popularity in medical image segmentation due to its ability to reduce reliance on image annotation. A typical approach in semi-supervised learning is to select reliable predictions as pseudo-labels and eliminate unreliable predictions. Contrastive learning helps prevent the insufficient utilization of unreliable predictions, but neglecting the anatomical structure of medical images can lead to suboptimal optimization results. We propose a novel approach for semi-supervised liver segmentation using contrastive learning, which leverages unlabeled data and enhances the suitability of contrastive learning for liver segmentation. Contrastive learning helps prevent the inappropriate utilization of unreliable predictions, but neglecting the anatomical structure of medical images can lead to suboptimal optimization results. Therefore, we propose a semi-supervised contrastive learning method with local regions self-supervision (LRS2). On one side, we employ Shannon entropy to distinguish between reliable and unreliable predictions and reduce the dissimilarity between their representations within regional artificial units. Within each unit of the liver image, we strongly encourage unreliable predictions to acquire valuable information pertaining to the correct category by leveraging the representations of reliable predictions in their vicinity. On the other side, we introduce a dynamic reliability threshold based on the Shannon entropy of each prediction, gradually evaluating the confidence threshold of reliable predictions as predictive accuracy improves. After selecting reliable predictions, we sequentially apply erosion and dilation to refine them for better selection of qualified positive and negative samples. We evaluate our proposed method on abdominal CT images, including 131 images (train data: 77, validation data: 26, and testing data: 28) from 2017 ISBI Liver Tumors Segmentation Challenge. Our method obtains satisfactory performance in different proportion by exploiting the unreliable predictions. Compared with the result of VNet only under supervised settings (with 10, 30, 50, 70% and full labeled data), LRS2, respectively, brings an improvement of Dice coefficient by +6.11, +3.55, +4.43, and +2.25%, achieving Dice coefficients of 93.44, 93.31, 94.85, and 95.12%,respectively. In this study, we carefully select appropriate positive and negative samples from reliable regions, ensuring that anchor pixels within unreliable regions are correctly assigned to their respective categories. With a consideration of the anatomical structure present in CT images, we partition the image representations into regional units, enabling anchor pixels to capture more precise sample information. Extensive experiments confirm the effectiveness of ourmethod.

Learning with limited annotations: A survey on deep semi-supervised learning for medical image segmentation

Medical image segmentation is a fundamental and critical step in many image-guided clinical approaches. Recent success of deep learning-based segmentation methods usually relies on a large amount of labeled data, which is particularly difficult and costly to obtain, especially in the medical imaging domain where only experts can provide reliable and accurate annotations. Semi-supervised learning has emerged as an appealing strategy and been widely applied to medical image segmentation tasks to train deep models with limited annotations. In this paper, we present a comprehensive review of recently proposed semi-supervised learning methods for medical image segmentation and summarize both the technical novelties and empirical results. Furthermore, we analyze and discuss the limitations and several unsolved problems of existing approaches. We hope this review can inspire the research community to explore solutions to this challenge and further advance the field of medical image segmentation.

Evidential Pseudo-Label Ensemble for semi-supervised classification

Semi-supervised learning (SSL) focuses on effectively exploiting both a limited number of labeled data and relatively rich unlabeled data. Recent SSL researches have achieved great progress with pseudo-labeling. However, incorrect pseudo-label may cause severe error propagation, which has been widely recognized and termed as confirmation bias. Furthermore, vanilla ensemble strategies aiming at alleviating this issue will lead to serious conservative labeling due to the averaging property. Accordingly, they could not sufficiently explore the unlabeled data. In this paper, we proposed a novel SSL method termed Evidential Pseudo-Label Ensemble (EPLE) which aims to generate more accurate pseudo-labels with evidence support. Specifically, multiple networks with different augmented strategies are introduced to generate complementary evidence for unlabeled samples, and induce confident predictions with Dempster–Shafer theory (DST), which effectively addresses error propagation and confirmation bias. To enhance the consistency of predictions for the same samples with different noises, we employ multiple augmentation strategies for both weak and strong augmentations. We conduct experiments on standard semi-supervised learning datasets, including CIFAR-10/100, SVHN and STL-10. Extensive experiments show the superior performance of our proposed EPLE compared to state-of-the-art SSL models.

Uncertainty-guided cross learning via CNN and transformer for semi-supervised honeycomb lung lesion segmentation

Objective. Deep learning networks such as convolutional neural networks (CNN) and Transformer have shown excellent performance on the task of medical image segmentation, however, the usual problem with medical images is the lack of large-scale, high-quality pixel-level annotations, which is a very time-consuming and laborious task, and its further leads to compromised the performance of medical image segmentation under limited annotation conditions. Approach. In this paper, we propose a new semi-supervised learning method, uncertainty-guided cross learning, which uses a limited number of annotated samples along with a large number of unlabeled images to train the network. Specifically, we use two networks with different learning paradigms, CNN and Transformer, for cross learning, and use the prediction of one of them as a pseudo label to supervise the other, so that they can learn from each other, fully extract the local and global features of the images, and combine explicit and implicit consistency regularization constraints with pseudo label methods. On the other hand, we use epistemic uncertainty as a guiding message to encourage the model to learn high-certainty pixel information in high-confidence regions, and minimize the impact of erroneous pseudo labels on the overall learning process to improve the performance of semi-supervised segmentation methods. Main results. We conducted honeycomb lung lesion segmentation experiments using a honeycomb lung CT image dataset, and designed several sets of comparison experiments and ablation experiments to validate the effectiveness of our method. The final experimental results show that the Dice coefficient of our proposed method reaches 88.49% on the test set, and our method achieves state-of-the-art performance in honeycomb lung lesion segmentation compared to other semi-supervised learning methods. Significance. Our proposed method can effectively improve the accuracy of segmentation of honeycomb lung lesions, which provides an important reference for physicians in the diagnosis and treatment of this disease.

A General Model for Aggregating Annotations Across Simple, Complex, and Multi-Object Annotation Tasks

Human annotations are vital to supervised learning, yet annotators often disagree on the correct label, especially as annotation tasks increase in complexity. A common strategy to improve label quality is to ask multiple annotators to label the same item and then aggregate their labels. To date, many aggregation models have been proposed for simple categorical or numerical annotation tasks, but far less work has considered more complex annotation tasks, such as those involving open-ended, multivariate, or structured responses. Similarly, while a variety of bespoke models have been proposed for specific tasks, our work is the first we are aware of to introduce aggregation methods that generalize across many, diverse complex tasks, including sequence labeling, translation, syntactic parsing, ranking, bounding boxes, and keypoints. This generality is achieved by applying readily available task-specific distance functions, then devising a task-agnostic method to model these distances between labels, rather than the labels themselves. This article presents a unified treatment of our prior work on complex annotation modeling and extends that work with investigation of three new research questions. First, how do complex annotation task and dataset properties impact aggregation accuracy? Second, how should a task owner navigate the many modeling choices in order to maximize aggregation accuracy? Finally, what tests and diagnoses can verify that aggregation models are specified correctly for the given data? To understand how various factors impact accuracy and to inform model selection, we conduct large-scale simulation studies and broad experiments on real, complex datasets. Regarding testing, we introduce the concept of unit tests for aggregation models and present a suite of such tests to ensure that a given model is not mis-specified and exhibits expected behavior. Beyond investigating these research questions above, we discuss the foundational concept and nature of annotation complexity, present a new aggregation model as a conceptual bridge between traditional models and our own, and contribute a new general semisupervised learning method for complex label aggregation that outperforms prior work.

CDGAN-BERT: Adversarial constraint and diversity discriminator for semi-supervised text classification

Semi-supervised generative adversarial learning (SS-GAN) significantly improves the model’s performance with limited labeled data, especially on text classification. However, these SS-GAN-like works pay more attention to the more real properties of the generated samples and ignore their adversarial nature, which may prevent the model from learning more higher-quality data representations. And the existing single discriminator may produce discrimination ambiguity when discriminating against tasks with different discriminative properties (simultaneously identifying real or fake and classifying categories), which may harm the model’s classification performance. In this paper, we propose a novel CDGAN-BERT, an SS-GAN-based architecture with the adversarial constraint and a diversity discriminator. Specifically, CDGAN-BERT first focuses on adversarial constraint, which calculates the model’s intrinsic state representation with the reciprocal of Mean Squared Error (MSE), further keeping adversarial of the generated samples relative to the real data distribution. Then, a diversity discriminator is designed to improve the model’s classification performance by alleviating discriminative ambiguity, which refines tasks with different discriminative attributes by adding an additional authenticity discriminator. We validate our model on 6 text datasets and achieve significant improvements, especially with limited supervision. The experimental results show that our method outperforms or gets comparable results to other state-of-the-art semi-supervised learning methods on several datasets. Especially for the datasets of QC-Fine and QC-Coarse with limited labeled data, our CDGAN-BERT achieves the best average Micro-F1 scores of 72.5% and 92.037%, respectively.

A domain knowledge powered hybrid regularization strategy for semi-supervised breast cancer diagnosis

Semi-supervised learning has attracted much attention in medical image analysis as annotating medical images is substantially difficult. Existing semi-supervised learning methods mainly utilize some regularization strategies to encourage a network to make similar predictions for each sample under various perturbations (e.g. rotating and flipping). However, as these strategies are originally designed for natural images, they do not fully utilize the specific attributes in the medical domain. In this paper, we propose DK-HRS, a domain knowledge-powered hybrid regularization strategy for semi-supervised breast cancer diagnosis. More specifically, DK-HRS generates four types of perturbed samples, namely, domain knowledge-augmented samples, weakly and strongly-augmented samples, and virtual adversarial samples. These perturbed samples represent transformations of medical images from different aspects. DK-HRS utilizes the FixMatch, a popular consistency regularization method as the backbone, and integrates two additional modules: a virtual adversarial training module, and a domain knowledge-based contrastive learning module. Experimental results on four breast ultrasound datasets demonstrate that, by incorporating medical domain knowledge, DK-HRS achieves superior performance and outperforms some state-of-the-art semi-supervised methods by a large margin.