Search

Abstract: Active learning in semi-supervised classification involves introducing additional labels for unlabelled data to improve the accuracy of the underlying classifier. A challenge is to identify which points to label to best improve performance while limiting the number of new labels. “Model Change” active learning quantifies the resulting change incurred in the classifier by introducing the additional label(s). We pair this idea with graph-based semi-supervised learning (SSL) methods, that use the spectrum of the graph Laplacian matrix, which can be truncated to avoid prohibitively large computational and storage costs. We consider a family of convex loss functions for which the acquisition function can be efficiently approximated using the Laplace approximation of the posterior distribution. We show a variety of multiclass examples that illustrate improved performance over prior state-of-art.

Fault diagnosis is an indispensable basis for the collaborative maintenance in prognostic and health management. Most of existing data-driven fault diagnosis approaches are designed in the framework of supervised learning, which requires a large number of labelled samples. In this paper, a novel self-training semi-supervised deep learning (SSDL) approach is proposed to train a fault diagnosis model together with few labelled and abundant unlabelled samples. The addressed SSDL approach is realised by initialising a stacked sparse auto-encoder classifier using the labelled samples, and subsequently updating the classifier via sampling a few candidates with most reliable pseudo labels from the unlabelled samples step by step. Unlike the commonly used static sampling strategy in existing self-training semi-supervised frameworks, a gradually exploiting mechanism is proposed in SSDL to increase the number of selected pseudo-labelled candidates gradually. In addition, instead of using the prediction accuracy as the confidence estimation for pseudo-labels, a distance-based sampling criterion is designed to assign the label for each unlabelled sample by its nearest labelled sample based on their Euclidean distances in the deep feature space. The experimental results show that the proposed SSDL approach can achieve good prediction accuracy compared to other self-training semi-supervised learning algorithms. [ABSTRACT FROM AUTHOR]

Summary: The issue of fault diagnosis in power equipment is receiving increasing attention from scholars. Due to the important role played by bearings in power equipment, bearing faults have become the main factor causing the shutdown of wind turbines units. Therefore, this paper takes bearing equipment as an example for research. In order to solve the problem of insufficient and unbalanced fault sample data of wind turbines bearings, a fault diagnosis (FD) method based on variational autoencoder and semi‐supervised learning is proposed in this paper. Firstly, based on Label Propagation‐random forests (LP‐RFs) and a small number of labeled fault samples, a semi‐supervised learning algorithm is proposed to label the original data samples. Secondly, a small number of training samples are preprocessed by the variational autoencoder to reduce the imbalance of the fault samples. Then, the RFs‐based method is adopted to train the processed fault samples to obtain a mature FD classifier. Finally, the proposed method is applied to FD for bearings, and the results show that the proposed method can realize bearings fault diagnosis (BFD). And meanwhile, the proposed method can also be applied for fault diagnosis in power transmission and transformation systems. [ABSTRACT FROM AUTHOR]

Semi-supervised classification has gained widespread popularity because of their superior ability to handle unlabeled samples in practical problems. This paper has presented a novel estimation error-ranked LSSVM method with double Mahalanobis-kernel which is used for semi-supervised classification. The main point is to construct two Mahalanobis distances in Hilbert space to form double Mahalanobis-kernel by considering the relationship between the characteristics of two sorts of samples, so as to reduce the influence of non-informational dimensions. Furthermore, the implementation of the proposed method is required to solve the label security problem of unlabeled samples. The unlabeled sample with the minimum evaluated error is selected for labeling, which effectively ensures the accuracy of the unlabeled sample labeling. This method not only considers the similarity of sample features, but also focuses on the security of unlabeled samples. And based on the experimental results of four artificial data sets and several UCI data sets, it verifies the effectiveness of the semi-supervised method with double Mahalanobis-kernel. Especially considering the experimental results of five disease diagnosis data sets, it demonstrates the potential of the proposed semi-supervised classification method in medical diagnosis. [ABSTRACT FROM AUTHOR]

Face clustering has applications in organizing personal photo album, video understanding and automatic labeling of data for semi-supervised learning. Many existing methods cannot cluster millions of faces. They are either too slow, inaccurate, or need a lot memory. In our paper, we proposed a two stage unsupervised clustering algorithm which can cluster millions of faces in minutes. A rough clustering using greedy Transitive Closure (TC) algorithm to separate the easy to locate clusters, then a more precise non-greedy clustering algorithm is used to split the clusters into smaller clusters. We also developed a set of omni-supervised transformations that can produce multiple embeddings using a single trained model as if there are multiple models trained. These embeddings are combined using simple averaging and normalization. We carried out extensive experiments with multiple datasets of different sizes comparing with existing state of the art clustering algorithms to show that our clustering algorithm is robust to differences between datasets, efficient and outperforms existing methods. We also carried out further analysis on number of singleton clusters and variations of our model using different non-greedy clustering algorithms. We did trained our semi-supervised model using the cluster labels and shown that our clustering algorithm is effective for semi-supervised learning. [ABSTRACT FROM AUTHOR]

Electroencephalogram (EEG) signals are derived from the central nervous system and inherently difficult to camouflage, leading to the recent popularity of EEG-based emotion recognition. However, due to the non-stationary nature of EEG, inter-subject variabilities become obstacles for recognition models to well adapt to different subjects. In this paper, we propose a novel approach called semi-supervised bipartite graph construction with active EEG sample selection (SBGASS) for cross-subject emotion recognition, which offers two significant advantages. Firstly, SBGASS adaptively learns a bipartite graph to characterize the underlying relationships between labeled and unlabeled EEG samples, effectively implementing the semantic connection for samples from different subjects. Secondly, we employ active sample selection technique in this paper to reduce the impact of negative samples (outliers or noise in the data) on bipartite graph construction. Drawing from the experimental results with the SEED-IV data set, we have gained the following three insights. (1) SBGASS actively rejects negative labeled samples, which helps mitigate the impact of negative samples when constructing the optimal bipartite graph and improves the model performance. (2) Through the learned optimal bipartite graph in SBGASS, the transferability of labeled EEG samples is quantitatively analyzed, which exhibits a decreasing tendency as the distance between each labeled sample and the corresponding class centroid increases. (3) Besides the improved recognition accuracy, the spatial-frequency patterns in emotion recognition are investigated by the acquired projection matrix. [ABSTRACT FROM AUTHOR]

Metric learning focuses on finding similarities between data and aims to enlarge the distance between the samples with different labels. This work proposes a semi-supervised metric learning method based on the point-to-class structure of the labeled data, which is computationally less expensive, especially than using point-to-point structure. Specifically, the point-to-class structure is formulated into a new triplet constraint, which could narrow the distance of inner-class data and enlarge the distance of inter-class data simultaneously. Moreover, for measuring dissimilarity between different classes, weights are introduced into the triplet constraint and forms the weighted triplet constraint. Then, two kinds of regularizers such as spatial regularizer are rationally incorporated respectively in this model to mitigate the overfitting phenomenon and preserve the topological structure of the data. Furthermore, Riemannian gradient descent algorithm is adopted to solve the proposed model, since it can fully exploit the geometric structure of Riemannian manifolds and the proposed model can be regarded as a generalization of the unconstrained optimization problem in Euclidean space on Riemannian manifold. By introducing such solution strategy, the variables are constrained to a specific Riemannian manifold in each step of the iterative solution process, thereby enabling efficient and accurate model resolution. Finally, we conduct classification experiments on various data sets and compare the classification performance to state-of-the-art methods. The experimental results demonstrate that our proposed method has better performance in classification, especially for hyperspectral image data. [ABSTRACT FROM AUTHOR]

Deep learning-based methods for fast target segmentation of computed tomography (CT) imaging have become increasingly popular. The success of current deep learning methods usually depends on a large amount of labeled data. Labeling medical data is a time-consuming and laborious task. Therefore, this paper aims to enhance the segmentation of CT images by using a semi-supervised learning method. In order to utilize the valid information in unlabeled data, we design a semi-supervised network model for contrastive learning based on entropy constraints. We use CNN and Transformer to capture the image's local and global feature information, respectively. In addition, the pseudo-labels generated by the teacher networks are unreliable and will lead to degradation of the model performance if they are directly added to the training. Therefore, unreliable samples with high entropy values are discarded to avoid the model extracting the wrong features. In the student network, we also introduce the residual squeeze and excitation module to learn the connection between different channels of each layer feature to obtain better segmentation performance. We demonstrate the effectiveness of the proposed method on the COVID-19 CT public dataset. We mainly considered three evaluation metrics: DSC, HD95, and JC. Compared with several existing state-of-the-art semi-supervised methods, our method improves DSC by 2.3%, JC by 2.5%, and reduces HD95 by 1.9 mm. In this paper, a semi-supervised medical image segmentation method is designed by fusing CNN and Transformer and utilizing entropy-constrained contrastive learning loss, which improves the utilization of unlabeled medical images. [ABSTRACT FROM AUTHOR]

Semi-supervised learning (SSL) employs unlabeled data with limited labeled samples to enhance deep networks, but imbalance degrades performance due to biased pseudo-labels skewing decision boundaries. To address this challenge, we propose two optimization conditions inspired by our theoretical analysis. These conditions focus on aligning class distributions and representations. Additionally, we introduce a plug-and-play method called Basis Transformation based distribution alignment (BTDA) that efficiently aligns class distributions while considering inter-class relationships. BTDA mitigates the negative impact of biased pseudo-labels through basis transformation, which involves a learnable transition matrix. Extensive experiments demonstrate the effectiveness of integrating existing SSL methods with BTDA in image classification tasks with class imbalance. For example, BTDA achieves accuracy improvements ranging from 2.47 to 6.66% on CIFAR10-LT and SVHN-LT datasets, and a remarkable 10.95% improvement on the tail class, even under high imbalanced rates. Despite its simplicity, BTDA achieves state-of-the-art performance in SSL with class imbalance on representative datasets. [ABSTRACT FROM AUTHOR]

The Graph Attention Network (GAT) is a type of graph neural network (GNN) that uses attention mechanisms to weigh the importance of nodes' neighbors, demonstrating flexibility and power in representation learning. However, GAT and its variants still face common challenges in GNNs, such as over-smoothing and over-squashing. To address this, we propose Multi-Head Multi-Order Graph Attention Networks (MHMOGAT) as an enhanced GAT layer. MHMOGAT is built based on multi-head attention and adjacency matrices of different orders, aiming to expand the receptive field of GAT to effectively capture long-distance dependencies. Moreover, Bayesian optimization is employed to determine optimal hyperparameter combinations for different datasets. Experimental results on six prevailing datasets demonstrate that MHMOGAT improves GAT accuracy by approximately 2-5% across various datasets with different label rates, indicating its effectiveness. Additionally, MHMOGAT exhibits potential in handling large and complex graphs with low label rates. [ABSTRACT FROM AUTHOR]

After oil spills occur in the ocean, oil pollutants usually appear in the form of oil emulsions under the influence of hydrodynamics. Hyperspectral remote sensing technology, which provides abundant spectral information of ground objects, has the potential of fine-grained classification on the types of oil emulsions. Aiming at the practical applications of oil emulsion extraction in hyperspectral images (HSIs), this study proposes a semi-supervised model for oil emulsion identification by integrating an image segmentation algorithm with a deep-learning-based classification model. In the proposed approach, the training data were filtered from HSI using an image segmentation algorithm, based on which a 1-dimensional convolutional neural network (1D-CNN) was trained to identify oil emulsions in the HSI. The model was tested on the HSIs of Deepwater Horizon oil spills obtained by AVIRIS. The overall accuracy and standard performance measurements of the proposed model are higher than 94% on the extracted dataset. The results indicated that the proposed model achieved similar detection results on sea water as the supervised model, and even higher accuracies on oil emulsion type identification. As a semi-supervised model, it also avoids the lengthy and time-consuming data labelling and has the potential for operational oil emulsions extraction and quantification. [ABSTRACT FROM AUTHOR]

In the ever-evolving domain of food computing, named entity recognition (NER) presents transformative potential that extends far beyond mere word tagging in recipes. Its implications encompass intelligent recipe recommendations, health analysis, and personalization. Nevertheless, existing NER models in food computing encounter challenges stemming from variations in recipe input standards, limited annotations, and dataset quality. This article addresses the specific problem of ingredient NER and introduces two innovative models: SINERA, an efficient and robust model, and SINERAS, a semi-supervised variant that leverages a Gaussian Mixture Model (GMM) to learn from untagged ingredient list entries. To mitigate issues associated with data quality and availability in food computing, we introduce the ARTI dataset, a diverse and comprehensive repository of ingredient lines. Additionally, we identify and tackle a pervasive challenge—spurious correlations between entity positions and predictions. To address this, we propose a set of data augmentation rules tailored for food NER. Extensive evaluations conducted on the ARTI dataset and a revised TASTEset dataset underscore the performance of our models. They outperform several state-of-the-art benchmarks and rival the BERT model while maintaining smaller parameter sizes and reduced training times. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNNs) have revolutionized various fields through their ability to model complex patterns, yet their performance critically hinges on the availability of large-scale, accurately labeled datasets. The costly and time-consuming process of creating such datasets has motivated research into Learning with Noisy Labels (LNL), which aims to reduce reliance on perfect labels. This paper introduces CoFix, an innovative LNL framework that integrates sample selection with semi-supervised learning techniques to address the challenge of noisy labels in training. CoFix employs a Gaussian Mixture Model (GMM) to dynamically segment the training data into clean and noisy subsets, leveraging semi-supervised learning approaches on both. Inspired by the FixMatch algorithm, CoFix refines consistent regularization and pseudo-labeling strategies, enhancing augmentation strategies and temperature sharpening techniques. Additionally, CoFix explores label smoothing to augment the loss function, further refining model performance. Our experiments demonstrate CoFix’s superiority over state-of-the-art methods, achieving significant improvements in fewer training epochs, particularly in lower noise scenarios. The robustness and versatility of CoFix are evident through its consistent performance across various benchmark datasets and noise levels. The contributions of this paper include a novel LNL method with enhanced generalization capability, an investigation into the impact of label smoothing on the loss function, and extensive testing that confirms CoFix’s efficiency and adaptability to different noise levels. [ABSTRACT FROM AUTHOR]

Summary: Imbalanced samples are widespread, which impairs the generalization and fairness of models. Semi‐supervised learning can overcome the deficiency of rare labeled samples, but it is challenging to select high‐quality pseudo‐label data. Unlike discrete labels that can be matched one‐to‐one with points on a numerical axis, labels in regression tasks are consecutive and cannot be directly chosen. Besides, the distribution of unlabeled data is imbalanced, which easily leads to an imbalanced distribution of pseudo‐label data, exacerbating the imbalance in the semi‐supervised dataset. To solve this problem, this article proposes a semi‐supervised imbalanced regression network (SIRN), which consists of two components: A, designed to learn the relationship between features and labels (targets), and B, dedicated to learning the relationship between features and target deviations. To measure target deviations under imbalanced distribution, the target deviation function is introduced. To select continuous pseudo‐labels, the deviation matching strategy is designed. Furthermore, an adaptive selection function is developed to mitigate the risk of skewed distributions due to imbalanced pseudo‐label data. Finally, the effectiveness of the proposed method is validated through evaluations of two regression tasks. The results show a great reduction in predicted value error, particularly in few‐shot regions. This empirical evidence confirms the efficacy of our method in addressing the issue of imbalanced samples in regression tasks. [ABSTRACT FROM AUTHOR]

The detailed characterization of the pore structure in sandstone is pivotal for the assessment of reservoir properties and the efficiency of oil and gas exploration. Traditional fully supervised learning algorithms are limited in performance enhancement and require a substantial amount of accurately annotated data, which can be challenging to obtain. To address this, we introduce a semi-supervised framework with a U-Net backbone network. Our dataset was curated from 295 two-dimensional CT grayscale images, selected at intervals from nine 4 mm sandstone core samples. To augment the dataset, we employed StyleGAN2-ADA to generate a large number of images with a style akin to real sandstone images. This approach allowed us to generate pseudo-labels through semi-supervised learning, with only a small subset of the data being annotated. The accuracy of these pseudo-labels was validated using ensemble learning methods. The experimental results demonstrated a pixel accuracy of 0.9993, with a pore volume discrepancy of just 0.0035 compared to the actual annotated data. Furthermore, by reconstructing the three-dimensional pore structure of the sandstone, we have shown that the synthetic three-dimensional pores can effectively approximate the throat length distribution of the real sandstone pores and exhibit high precision in simulating throat shapes. [ABSTRACT FROM AUTHOR]

Low probability of intercept (LPI) radar signal recognition under low signal-to-noise ratio (SNR) is a challenging task within electronic reconnaissance systems, particularly when faced with scarce labeled data and limited resources. In this paper, we introduce an LPI radar signal recognition method based on a semi-supervised Support Vector Machine (SVM). First, we utilize the Multi-Synchrosqueezing Transform (MSST) to obtain the time–frequency images of radar signals and undergo the necessary preprocessing operations. Then, the image features are extracted via Discrete Wavelet Transform (DWT), and the feature dimension is reduced by the principal component analysis (PCA). Finally, the dimensionality reduction features are input into the semi-supervised SVM to complete the classification and recognition of LPI radar signals. The experimental results demonstrate that the proposed method achieves high recognition accuracy at low SNR. When the SNR is −6 dB, its recognition accuracy reaches almost 100%. [ABSTRACT FROM AUTHOR]

The development of mobile and on the edge applications that embed deep convolutional neural models has the potential to revolutionise healthcare. However, most deep learning models require computational resources that are not available in smartphones or edge devices; an issue that can be faced by means of compact models that require less resources than standard deep learning models. The problem with such models is that they are, at least usually, less accurate than bigger models. We propose to address the accuracy limitation of compact networks with the application of semi-supervised learning techniques. In particular, we perform a thorough comparison of self-training methods, consistency regularisation techniques, and quantization techniques. We present a thorough analysis for the results obtained by combining 11 compact networks and 6 semi-supervised processes when applied to 10 biomedical datasets. Namely, combining semi-supervised methods and compact networks, we can create compact models that are not only as accurate as standard size models, but also faster and lighter. In addition, we have developed a Python library to facilitate the combination of compact networks and semi-supervised learning methods to tackle image classification tasks. [ABSTRACT FROM AUTHOR]

Human action recognition (HAR) technology based on radar signals has garnered significant attention from both industry and academia due to its exceptional privacy-preserving capabilities, noncontact sensing characteristics, and insensitivity to lighting conditions. However, the scarcity of accurately labeled human radar data poses a significant challenge in meeting the demand for large-scale training datasets required by deep model-based HAR technology, thus substantially impeding technological advancements in this field. To address this issue, a semi-supervised learning algorithm, MF-Match, is proposed in this paper. This algorithm computes pseudo-labels for larger-scale unsupervised radar data, enabling the model to extract embedded human behavioral information and enhance the accuracy of HAR algorithms. Furthermore, the method incorporates contrastive learning principles to improve the quality of model-generated pseudo-labels and mitigate the impact of mislabeled pseudo-labels on recognition performance. Experimental results demonstrate that this method achieves action recognition accuracies of 86.69% and 91.48% on two widely used radar spectrum datasets, respectively, utilizing only 10% labeled data, thereby validating the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Ship detection from synthetic aperture radar (SAR) imagery is crucial for various fields in real-world applications. Numerous deep learning-based detectors have been investigated for SAR ship detection, which requires a substantial amount of labeled data for training. However, SAR data annotation is time-consuming and demands specialized expertise, resulting in deep learning-based SAR ship detectors struggling due to a lack of annotations. With limited labeled data, semi-supervised learning is a popular approach for boosting detection performance by excavating valuable information from unlabeled data. In this paper, a semi-supervised SAR ship detection network is proposed, termed a Multi-Teacher Dempster-Shafer Evidence Fusion Net-work (MTDSEFN). The MTDSEFN is an enhanced framework based on the basic teacher–student skeleton frame, comprising two branches: the Teacher Group (TG) and the Agency Teacher (AT). The TG utilizes multiple teachers to generate pseudo-labels for different augmentation versions of unlabeled samples, which are then refined to obtain high-quality pseudo-labels by using Dempster-Shafer (D-S) fusion. The AT not only serves to deliver weights of its own teacher to the TG at the end of each epoch but also updates its own weights after each iteration, enabling the model to effectively learn rich information from unlabeled data. The combination of TG and AT guarantees both reliable pseudo-label generation and a comprehensive diversity of learning information from numerous unlabeled samples. Extensive experiments were performed on two public SAR ship datasets, and the results demonstrated the effectiveness and superiority of the proposed approach. [ABSTRACT FROM AUTHOR]

Pavement cracks are crucial indicators for assessing the structural health of asphalt roads. Existing automated crack detection models depend on large quantities of precisely annotated crack sample data. The irregular morphology of cracks makes manual annotation time-consuming and costly, thereby posing challenges to the practical application of these models. This study proposes a pavement crack image detection method integrating fractal dimension analysis and semi-supervised learning. It identifies the self-similarity characteristics within the crack regions by analyzing pavement crack images and using fractal dimensions to preliminarily determine the candidate crack regions. The Crack Similarity Learning Network (CrackSL-Net) is then employed to learn the semantic similarity of crack image regions. Semi-supervised learning facilitates automatic crack detection by combining a small amount of labeled data with a large volume of unlabeled image data. Comparative experiments are conducted on two public pavement crack datasets against the HED, U-Net, and RCF models to comprehensively evaluate the performance of the proposed method. The results indicate that, with a 50% annotation ratio, the proposed method achieves high-precision crack detection, with an intersection over union (IoU) exceeding 0.84, which is close to that of U-Net. Visual analysis of the detection results confirms the method's effectiveness in identifying cracks in complex environments. [ABSTRACT FROM AUTHOR]

Featured Application: This study focuses on the segmentation of 3D medical images. Due to the difficulty in obtaining labels for medical images, it is challenging to achieve effective segmentation using fully supervised networks. Therefore, semi-supervised networks have become a better choice. To enhance the segmentation performance of semi-supervised learning methods on the left-atrium database, this paper proposes a triple consistency segmentation network based on squeeze–excitation (SETC-Net). Our semi-supervised medical image segmentation network achieves state-of-the-art segmentation performance on the left-atrium database. Convolutional neural networks (CNNs) have achieved remarkable success in fully supervised medical image segmentation tasks. However, the acquisition of large quantities of homogeneous labeled data is challenging, making semi-supervised training methods that rely on a small amount of labeled data and pseudo-labels increasingly popular in recent years. Most existing semi-supervised learning methods, however, underestimate the importance of the unlabeled regions during training. This paper posits that these regions may contain crucial information for minimizing the model's uncertainty prediction. To enhance the segmentation performance of the left-atrium database, this paper proposes a triple consistency segmentation network based on the squeeze-and-excitation mechanism (SETC-Net). Specifically, the paper constructs a symmetric architectural unit called SEConv, which adaptively recalibrates the feature responses in the channel direction by modeling the inter-channel correlations. This allows the network to adaptively weigh each channel according to the task's needs, thereby emphasizing or suppressing different feature channels. Moreover, SETC-Net is composed of an encoder and three slightly different decoders, which convert the prediction discrepancies among the three decoders into unsupervised loss through a constructed iterative pseudo-labeling scheme, thus encouraging consistent and low-entropy predictions. This allows the model to gradually capture generalized features from these challenging unmarked regions. We evaluated the proposed SETC-Net on the public left-atrium (LA) database. The proposed method achieved an excellent Dice score of 91.14% using only 20% of the labeled data. The experiments demonstrate that the proposed SETC-Net outperforms seven current semi-supervised methods in left-atrium segmentation and is one of the best semi-supervised segmentation methods on the LA database. [ABSTRACT FROM AUTHOR]

With increasing wafer sizes and diversifying die patterns, automated optical inspection (AOI) is progressively replacing traditional visual inspection (VI) for wafer defect detection. Yet, the defect classification efficacy of current AOI systems in our case company is not optimal. This limitation is due to the algorithms' reliance on expertly designed features, reducing adaptability across various product models. Additionally, the limited time available for operators to annotate defect samples restricts learning potential. Our study introduces a novel hybrid self-training algorithm, leveraging semi-supervised learning that integrates pseudo-labeling, noisy student, curriculum labeling, and the Taguchi method. This approach enables classifiers to autonomously integrate information from unlabeled data, bypassing the need for feature extraction, even with scarcely labeled data. Our experiments on a small-scale set show that with 25% and 50% labeled data, the method achieves over 92% accuracy. Remarkably, with only 10% labeled data, our hybrid method surpasses the supervised DenseNet classifier by over 20%, achieving more than 82% accuracy. On a large-scale set, the hybrid method consistently outperforms other approaches, achieving up to 88.75%, 86.31%, and 83.61% accuracy with 50%, 25%, and 10% labeled data. Further experiments confirm our method's consistent superiority, highlighting its potential for high classification accuracy in limited-data scenarios. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) have achieved breakthrough progress in various fields, largely owing to the support of large-scale datasets with manually annotated labels. However, obtaining such datasets is costly and time-consuming, making high-quality annotation a challenging task. In this work, we propose an improved noisy sample selection method, termed "sample selection framework", based on a mixup loss and recalibration strategy (SMR). This framework enhances the robustness and generalization abilities of models. First, we introduce a robust mixup loss function to pre-train two models with identical structures separately. This approach avoids additional hyperparameter adjustments and reduces the need for prior knowledge of noise types. Additionally, we use a Gaussian Mixture Model (GMM) to divide the entire training set into labeled and unlabeled subsets, followed by robust training using semi-supervised learning (SSL) techniques. Furthermore, we propose a recalibration strategy based on cross-entropy (CE) loss to prevent the models from converging to local optima during the SSL process, thus further improving performance. Ablation experiments on CIFAR-10 with 50% symmetric noise and 40% asymmetric noise demonstrate that the two modules introduced in this paper improve the accuracy of the baseline (i.e., DivideMix) by 1.5% and 0.5%, respectively. Moreover, the experimental results on multiple benchmark datasets demonstrate that our proposed method effectively mitigates the impact of noisy labels and significantly enhances the performance of DNNs on noisy datasets. For instance, on the WebVision dataset, our method improves the top-1 accuracy by 0.7% and 2.4% compared to the baseline method. [ABSTRACT FROM AUTHOR]

This study presents the development and validation of a robust semi-supervised learning framework specifically designed for the automated segmentation and classification of sandstone thin section images from the Yanchang Formation in the Ordos Basin. Traditional geological image analysis methods encounter significant challenges due to the labor-intensive and error-prone nature of manual labeling, compounded by the diversity and complexity of rock thin sections. Our approach addresses these challenges by integrating the GL-SLIC algorithm, which combines Gabor filters and Local Binary Patterns for effective superpixel segmentation, laying the groundwork for advanced component identification. The primary innovation of this research is the semi-supervised learning model that utilizes a limited set of manually labeled samples to generate high-confidence pseudo labels, thereby significantly expanding the training dataset. This methodology effectively tackles the critical challenge of insufficient labeled data in geological image analysis, enhancing the model's generalization capability from minimal initial input. Our framework improves segmentation accuracy by closely aligning superpixels with the intricate boundaries of mineral grains and pores. Additionally, it achieves substantial improvements in classification accuracy across various rock types, reaching up to 96.3% in testing scenarios. This semi-supervised approach represents a significant advancement in computational geology, providing a scalable and efficient solution for detailed petrographic analysis. It not only enhances the accuracy and efficiency of geological interpretations but also supports broader hydrocarbon exploration efforts. [ABSTRACT FROM AUTHOR]

Although modulation classification with deep learning has been widely explored, this is challenging when the training data is limited. In this paper, we meet this challenge by data augmentation based on a semi-supervised generative model, named semi-supervised variational auto-encoder GAN (SS-VAEGAN). The proposed model has two novel aspects: first, the ensemble prediction technique is presented to establish a semi-supervised framework, and thus few labeled data and abundant unlabeled data can be used simultaneously to train the generative model; second, a modified feature matching strategy is utilized to ensure the diversity of the generated samples and contribute to the final classification under limited labeled data. With the aim of the constellation diagram with density and the ResNet-18 network, we have worked out a complete solution of modulation classification with data augmentation using SS-VAEGAN. In the experiments, we investigate the classification accuracy and robustness of the proposed data augmentation method and then compare the proposed SS-VAEGAN with other deep generative models. It is shown that our method can achieve the classification accuracy of 0.968 in the extreme absence of training data. [ABSTRACT FROM AUTHOR]

When sampling data of specific classes (i.e., known classes) for a scientific task, collectors may encounter unknown classes (i.e., novel classes). Since these novel classes might be valuable for future research, collectors will also sample them and assign them to several clusters with the help of known-class data. This assigning process is known as novel class discovery (NCD). However, category confusion is common in the sampling process and may make the NCD unreliable. To tackle this problem, this paper introduces a new and more realistic setting, where collectors may misidentify known classes and even confuse known classes with novel classes—we name it NCD under unreliable sampling (NUSA). We find that NUSA will empirically degrade existing NCD methods if taking no care of sampling errors. To handle NUSA, we propose an effective solution, named hidden-prototype-based discovery network (HPDN): (1) we try to obtain relatively clean data representations even with the confusedly sampled data; (2) we propose a mini-batch K-means variant for robust clustering, alleviating the negative impact of residual errors embedded in the representations by detaching the noisy supervision timely. Experiments demonstrate that, under NUSA, HPDN significantly outperforms competitive baselines (e.g., 6 % more than the best baseline on CIFAR-10) and remains robust when encountering serious sampling errors. [ABSTRACT FROM AUTHOR]

In response to the scarcity of defect samples in industrial applications, this paper proposed a method for segmenting surface defects in metal workpieces with only a minimal number of required defect samples. The method combined image generation techniques with a semi-supervised learning strategy. It utilized small-sized defect patches, and extracted from a minimal number of defect images to train a defect generation model. Subsequently, the method integrated these generated defect images into normal images to facilitate data augmentation. Additionally, the method applied a semi-supervised learning strategy to train the segmentation network, aiming to mitigate the adverse effects of differences between generated and real data distributions. The experimental phase involved conducting tests on a real-world computer vision detection system for metal workpieces. The results demonstrate that the semi-supervised training strategy significantly enhances the segmentation model's generalization ability to real defects. The method achieves high segmentation accuracy using only five defect sample images. [ABSTRACT FROM AUTHOR]

This paper introduces a method aimed at diagnosing the presence or absence of lesions by detecting anomalies in Chest X-ray images. The proposed approach is based on the PatchCore anomaly detection method, which extracts a feature vector containing location information of an image patch from normal image data and calculates the anomaly distance from the normal vector. However, applying PatchCore directly to medical image processing presents challenges due to the possibility of diseases occurring only in specific organs and the presence of image noise unrelated to lesions. In this study, we present an image alignment method that utilizes affine transformation parameter prediction to standardize already captured X-ray images into a specific composition. Additionally, we introduce a region-specific abnormality detection method that requires affine-transformed chest X-ray images. Furthermore, we propose a method to enhance application efficiency and performance through feature map hard masking. The experimental results demonstrate that our proposed approach achieved a maximum AUROC (Area Under the Receiver Operating Characteristic) of 0.774. Compared to a previous study conducted on the same dataset, our method shows a 6.9% higher performance and improved accuracy. [ABSTRACT FROM AUTHOR]

Magnetic resonance imaging (MRI) is a ubiquitous medical imaging technology with applications in disease diagnostics, intervention, and treatment planning. Accurate MRI segmentation is critical for diagnosing abnormalities, monitoring diseases, and deciding on a course of treatment. With the advent of advanced deep learning frameworks, fully automated and accurate MRI segmentation is advancing. Traditional supervised deep learning techniques have advanced tremendously, reaching clinical‐level accuracy in the field of segmentation. However, these algorithms still require a large amount of annotated data, which is oftentimes unavailable or impractical. One way to circumvent this issue is to utilize algorithms that exploit a limited amount of labeled data. This paper aims to review such state‐of‐the‐art algorithms that use a limited number of annotated samples. We explain the fundamental principles of self‐supervised learning, generative models, few‐shot learning, and semi‐supervised learning and summarize their applications in cardiac, abdomen, and brain MRI segmentation. Throughout this review, we highlight algorithms that can be employed based on the quantity of annotated data available. We also present a comprehensive list of notable publicly available MRI segmentation datasets. To conclude, we discuss possible future directions of the field—including emerging algorithms, such as contrastive language‐image pretraining, and potential combinations across the methods discussed—that can further increase the efficacy of image segmentation with limited labels. [ABSTRACT FROM AUTHOR]

Semi-supervised learning is a promising approach to dealing with the problem of insufficient labeled data. Recent methods grouped into paradigms of consistency regularization and pseudo-labeling have outstanding performances on image data, but achieve limited improvements when employed for processing textual information, due to the neglect of the discrete nature of textual information and the lack of high-quality text augmentation transformation means. In this paper, we propose the novel SeqMatch method. It can automatically perceive abnormal model states caused by anomalous data obtained by text augmentations and reduce their interferences and instead leverages normal ones to improve the effectiveness of consistency regularization. And it generates hard artificial pseudo-labels to enable the model to be efficiently updated and optimized toward low entropy. We also design several much stronger well-organized text augmentation transformation pipelines to increase the divergence between two views of unlabeled discrete textual sequences, thus enabling the model to learn more knowledge from the alignment. Extensive comparative experimental results show that our SeqMatch outperforms previous methods on three widely used benchmarks significantly. In particular, SeqMatch can achieve a maximum performance improvement of 16.4% compared to purely supervised training when provided with a minimal number of labeled examples. [ABSTRACT FROM AUTHOR]

The self-labeled methods have been favored by scholars in semi-supervised classification. Mislabeling is a great challenge for self-labeled methods and one of the reasons for mislabeling is that high-confidence unlabeled samples are found by mistake. While multiple variations of self-labeled methods have been developed, most existing strategies for finding high-confidence unlabeled samples heavily rely on specific assumptions. To solve the above issue, a binary particle swarm optimization-based self-labeled method (BPSO-SLM) is proposed and includes the following iterative self-labeled process: (a) A given classifier is trained on the set of labeled data; (b) The binary particle swarm optimization-based sample subspace optimization (BPSOSSO) is innovatively proposed to help BPSO-SLM find high-confidence unlabeled samples and low-confidence unlabeled samples from the set of unlabeled data; (c) The trained classifier is used to predict found high-confidence unlabeled samples; (d) High-confidence samples with pseudo labels are added to the set of labeled data, while low-confidence samples are returned to the set of unlabeled data and will be predicted again in the next iteration. The above process repeats until no high-confidence samples are found. After that, BPSO-SLM outputs the trained classifier during the iterative self-labeled process. The main characteristic of BPSO-SLM is that the strategy of finding high-confidence unlabeled samples makes little or no specific hypothesis about the geometric, distribution, and others of the selected high-confidence unlabeled samples. Intensive experiments on benchmark data prove that BPSO-SLM outperforms 8 state-of-the-art self-labeled methods in classification accuracy, Marco F-measure, and labeling error rate with various ratios of labeled data. [ABSTRACT FROM AUTHOR]

Depth completion is a critical task for extensive applications such as 3D reconstruction and object detection. Recent semi-supervised depth completion techniques based on Stereo-LiDAR fusion has gradually attracted attention due to its ability to realize semi-supervised training under the weakly supervised constraints provided by the sparse LiDAR points as well as the view synthesis consistency provided by the stereo images. However, among these methods, noisy label as a conventional supervised signal for semi-supervised learning have not been taken seriously. To this end, we first propose a semi-supervised learning framework called Recursive Noise Label Learning (RNLL), which is able to maximize the potential of noisy label learning by establishing a recursive noisy label optimization and model learning mechanism. Second, based on the proposed RNLL framework, we come up with a novel semi-supervised depth completion model, UAMD-Net-RNLL, which integrates the CCNN model to accomplish the confidence measure of depth prediction on the basis of the original UAMD-Net model, and realizes the noise filtering of the noisy depth maps and adaptive loss computation by constructing the confidence-guided noisy label loss function. Finally, for the noisy label and photometric consistency loss constraints in multiple rounds of training, we utilize dynamic loss weights to describe their importance changes, and realize the continuous improvement of the prediction performance of the semi-supervised depth completion model. Extensive experimental results conducted on KITTI and DrivingStereo depth completion datasets prove that our technique achieves large performance gain, clearly outperforming competing methods and setting a new state-of-the-art for semi-supervised depth completion. [ABSTRACT FROM AUTHOR]

Federated Learning (FL) provides a novel paradigm for privacy-preserving machine learning, enabling multiple clients to collaborate on model training without sharing private data. To handle multi-source heterogeneous data, Vertical Federated Learning (VFL) has been extensively investigated. However, in the context of VFL, the label information tends to be kept in one authoritative client and is very limited. This poses two challenges for model training in the VFL scenario. On the one hand, a small number of labels cannot guarantee to train a well VFL model with informative network parameters, resulting in unclear boundaries for classification decisions. On the other hand, the large amount of unlabeled data is dominant and should not be discounted, and it is worthwhile to focus on how to leverage them to improve representation modeling capabilities. To address the preceding two challenges, we first introduce supervised contrastive loss to enhance the intra-class aggregation and inter-class estrangement, which is to deeply explore label information and improve the effectiveness of downstream classification tasks. Then, for unlabeled data, we introduce a pseudo-label-guided consistency mechanism to induce the classification results coherent across clients, which allows the representations learned by local networks to absorb the knowledge from other clients, and alleviates the disagreement between different clients for classification tasks. We conduct sufficient experiments on four commonly used datasets, and the experimental results demonstrate that our method is superior to the state-of-the-art methods, especially in the low-label rate scenario, and the improvement becomes more significant. [ABSTRACT FROM AUTHOR]

Semi-supervised learning(SSL) can substantially improve the performance of deep neural networks by utilizing unlabeled data when labeled data is scarce. The state-of-the-art(SOTA) semi-supervised algorithms implicitly assume that the class distribution of labeled datasets and unlabeled datasets are balanced, which means the different classes have the same numbers of training samples. However, they can hardly perform well on minority classes when the class distribution of training data is imbalanced. Recent work has found several ways to decrease the degeneration of semi-supervised learning models in class-imbalanced learning. In this article, we comprehensively review class-imbalanced semi-supervised learning (CISSL), starting with an introduction to this field, followed by a realistic evaluation of existing class-imbalanced semi-supervised learning algorithms and a brief summary of them. [ABSTRACT FROM AUTHOR]

Pseudo-labeling methods are popular in semi-supervised learning (SSL). Their performance heavily relies on a proper threshold to generate hard labels for unlabeled data. To this end, most existing studies resort to a manually pre-specified function to adjust the threshold, which, however, requires prior knowledge and suffers from the scalability issue. In this paper, we propose a novel method named Meta-Threshold, which learns a dynamic confidence threshold for each unlabeled instance and does not require extra hyperparameters except a learning rate. Specifically, the instance-level confidence threshold is automatically learned by an extra network in a meta-learning manner. Considering limited labeled data as meta-data, the overall training objective of the classifier network and the meta-net can be formulated as a nested optimization problem that can be solved by a bi-level optimization scheme. Furthermore, by replacing the indicator function existed in the pseudo-labeling with a surrogate function, we theoretically provide the convergence of our training procedure, while discussing the training complexity and proposing a strategy to reduce its time cost. Extensive experiments and analyses demonstrate the effectiveness of our method on both typical and imbalanced SSL tasks. [ABSTRACT FROM AUTHOR]

Copyright of Systems Engineering & Electronics is the property of Journal of Systems Engineering & Electronics Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Due to the lack of sufficient valid labeled data and severe channel fading, the recognition of various underwater acoustic (UWA) communication modulation types still faces significant challenges. In this paper, we propose a lightweight UWA communication type recognition network based on semi-supervised learning, named the SSL-LRN. In the SSL-LRN, a mean teacher–student mechanism is developed to improve learning performance by averaging the weights of multiple models, thereby improving recognition accuracy for insufficiently labeled data. The SSL-LRN employs techniques such as quantization and small convolutional kernels to reduce floating-point operations (FLOPs), enabling its deployment on underwater mobile nodes. To mitigate the performance loss caused by quantization, the SSL-LRN adopts a channel expansion module to optimize the neuron distribution. It also employs an attention mechanism to enhance the recognition robustness for frequency-selective-fading channels. Pool and lake experiments demonstrate that the framework effectively recognizes most modulation types, achieving a more than 5% increase in recognition accuracy at a 0 dB signal-to-noise ratio (SNRs) while reducing FLOPs by 84.9% compared with baseline algorithms. Even with only 10% labeled data, the performance of the SSL-LRN approaches that of the fully supervised LRN algorithm. [ABSTRACT FROM AUTHOR]

While significant progress has been made in image dehazing techniques, the lack of large-scale labeled datasets remains one of the limiting factors for enhancing the performance of image dehazing algorithms. Therefore, based on the mean teacher model, we propose a semi-supervised dehazing framework with a dual-teacher-student (DTS) architecture. DTS is composed of a pretrained teacher network (P-teacher), a mean teacher network (M-teacher), and a student network. The P-teacher facilitates the student network in learning intermediate layer features that resemble haze-free images through knowledge transfer. The M-teacher guides the student network in image dehazing in unsupervised manner. The P-teacher, M-teacher, and student networks share the same network architecture known as the multiscale feature fusion attention-enhanced network (MFFA-Net). The MFFA-Net consists of a multiscale feature fusion network (MFF-Net) and an attention network (A-Net). The MFF-Net is responsible for fusing features from different levels. The A-Net is capable of compensating for information loss during downsampling in the MFF-Net and dynamically adjusting the focus on different regions. Extensive experimental results demonstrate that the dehazing method proposed in this paper outperforms several state-of-the-art algorithms on multiple datasets.The code has been released on https://github.com/houqianwen/MFFA-Net. [ABSTRACT FROM AUTHOR]

Currently, although semi-supervised image segmentation has achieved significant success in many aspects, further improvement in segmentation accuracy is necessary for practical applications. Additionally, there are fewer networks specifically designed for segmenting 3D images compared to those for 2D images, and their performance is notably inferior. To enhance the efficiency of network training, various attention mechanisms have been integrated into network models. However, these networks have not effectively extracted all the useful spatial or channel information. Particularly for 3D medical images, which contain rich spatial and channel information with tightly interconnected relationships between them, there remains a wealth of spatial and channel-specific information waiting to be explored and utilized. This paper proposes a bidirectional and efficient attention parallel network (BEAP-Net). Specifically, we introduce two modules: Supreme Channel Attention (SCA) and Parallel Spatial Attention (PSA). These modules aim to extract more spatial and channel-specific feature information and effectively utilize it. We combine the principles of consistency training and entropy regularization to enable mutual learning among sub-models. We evaluate the proposed BEAP-Net on two public 3D medical datasets, LA and Pancreas. The network outperforms the current state of the art in eight algorithms and is better suited for 3D medical images. It achieves the new best semi-supervised segmentation performance on the LA database. Ablation studies further validate the effectiveness of each component of the proposed model. Moreover, the SCA and PSA modules proposed can be seamlessly integrated into other 3D medical image segmentation networks to yield significant performance gains. [ABSTRACT FROM AUTHOR]

Cardiotocography (CTG) measurements are critical for assessing fetal wellbeing during monitoring, and accurate assessment requires well-traceable CTG signals. The current FHR calculation algorithm, based on autocorrelation to Doppler ultrasound (DUS) signals, often results in periods of loss owing to its inability to differentiate signals. We hypothesized that classifying DUS signals by type could be a solution and proposed that an artificial intelligence (AI)-based approach could be used for classification. However, limited studies have incorporated the use of AI for DUS signals because of the limited data availability. Therefore, this study focused on evaluating the effectiveness of semi-supervised learning in enhancing classification accuracy, even in limited datasets, for DUS signals. Data comprising fetal heartbeat, artifacts, and two other categories were created from non-stress tests and labor DUS signals. With labeled and unlabeled data totaling 9,600 and 48,000 data points, respectively, the semi-supervised learning model consistently outperformed the supervised learning model, achieving an average classification accuracy of 80.9%. The preliminary findings indicate that applying semi-supervised learning to the development of AI models using DUS signals can achieve high generalization accuracy and reduce the effort. This approach may enhance the quality of fetal monitoring. [ABSTRACT FROM AUTHOR]

This paper presents an on-device semi-supervised human activity detection system that can learn and predict human activity patterns in real time. The clinical objective is to monitor and detect the unhealthy sedentary lifestyle of a user. The proposed semi-supervised learning (SSL) framework uses sparsely labelled user activity events acquired from Inertial Measurement Unit sensors installed as wearable devices. The proposed cluster-based learning model in this approach is trained with data from the same target user, thus preserving data privacy while providing personalized activity detection services. Two different cluster labelling strategies, namely, population-based and distance-based strategies, are employed to achieve the desired classification performance. The proposed system is shown to be highly accurate and computationally efficient for different algorithmic parameters, which is relevant in the context of limited computing resources on typical wearable devices. Extensive experimentation and simulation study have been conducted on multi-user human activity data from the public domain in order to analyze the trade-off between classification accuracy and computation complexity of the proposed learning paradigm with different algorithmic hyper-parameters. With 4.17 h of training time for 8000 activity episodes, the proposed SSL approach consumes at most 20 KB of CPU memory space, while providing a maximum accuracy of 90% and 100% classification rates. [ABSTRACT FROM AUTHOR]

Data, in deep learning (DL), are crucial to detect ships in synthetic aperture radar (SAR) images. However, SAR image annotation limitations hinder DL-based SAR ship detection. A novel data-selection method and teacher–student model are proposed in this paper to effectively leverage sparse labeled data and improve SAR ship detection performance, based on the semi-supervised oriented object-detection (SOOD) framework. More specifically, we firstly propose a SAR data-scoring method based on fuzzy comprehensive evaluation (FCE), and discuss the relationship between the score distribution of labeled data and detection performance. A refined data selector (RDS) is then designed to adaptively obtain reasonable data for model training without any labeling information. Lastly, a Gaussian Wasserstein distance (GWD) and an orientation-angle deviation weighting (ODW) loss are introduced to mitigate the impact of strong scattering points on bounding box regression and dynamically adjusting the consistency of pseudo-label prediction pairs during the model training process, respectively. The experiments results on four open datasets have demonstrated that our proposed method can achieve better SAR ship detection performances on low-proportion labeled datasets, compared to some existing methods. Therefore, our proposed method can effectively and efficiently reduce the burden of SAR ship data labeling and improve detection capacities as much as possible. [ABSTRACT FROM AUTHOR]

Identifying calcifications in mammograms is crucial for early breast cancer detection, and semi-supervised learning, which utilizes a small dataset for supervised learning combined with deep learning, is anticipated to be an effective approach for automating this identification process. This study explored the impact of semi-supervised learning on identifying mammographic calcifications by including 712 mammographic images from 252 patients in public datasets. Initially, 212 mammogram images were segmented into patches and classified visually for calcification presence. A subset of these patches, derived from 169 mammogram images, was used to train a ResNet50-based classifier. The classifier was evaluated using patches generated from 43 mammograms as a test data set. Additionally, 500 more mammogram images were processed into patches and analyzed using the trained ResNet50 model, with semi-supervised learning applied to patches exceeding certain classification probabilities. This process aimed to enhance the classifier's accuracy and achieve improvements over the initial model. The findings indicated that semi-supervised learning significantly benefits the accuracy of calcification detection in mammography, underscoring its utility in enhancing diagnostic methodologies. [ABSTRACT FROM AUTHOR]

Clustering algorithms play a crucial role in early warning cybersecurity systems. They allow for the detection of new attack patterns and anomalies and enhance system performance. This paper discusses the problem of clustering data collected by a distributed system of network honeypots. In the proposed approach, when a network flow matches an attack signature, an appropriate label is assigned to it. This enables the use of semi-supervised learning algorithms and improves the quality of clustering results. The article compares the results of learning algorithms conducted with and without partial supervision, particularly non-negative matrix factorization and semi-supervised non-negative matrix factorization. Our results confirm the positive impact of labeling a portion of flows on the quality of clustering. [ABSTRACT FROM AUTHOR]

With the rapid development of Industry 4.0, the importance of cyber security for industrial control systems has become increasingly prominent. The complexity and diversity of industrial control systems result in data with high dimensionality and strong correlation, posing significant challenges in obtaining labeled data. However, current intrusion detection methods often demand large amounts of labeled data for effective training. To address this limitation, this paper proposes a semi-supervised anomaly detection framework, called SFSD, which leverages feature selection and deviation networks to detect anomalies in industrial control systems. Specifically, we introduce a feature selection algorithm (IG-PCA) that utilizes information gain and principal component analysis to reduce the dimensionality of features in industrial control data by eliminating redundant features. Then, we propose a semi-supervised learning method based on an improved deviation network, which utilizes an anomaly scoring network to learn end-to-end anomaly scores for the training data, thus assigning anomaly scores to each training data. Finally, using a limited amount of anomaly-labeled data, we design a specific deviation loss function to optimize the anomaly scoring network, enabling a significant score bias between positive and negative samples. Experimental results demonstrate that the proposed SFSD outperforms existing semi-supervised anomaly detection frameworks by improving the accuracy and detection rate by an average of 1–2%. Moreover, SFSD requires less training time compared to existing frameworks, resulting in a training time reduction of approximately 10% or more. [ABSTRACT FROM AUTHOR]

The switch machine, an essential element of railway infrastructure, is crucial in maintaining the safety of railway operations. Traditional methods for fault diagnosis are constrained by their dependence on extensive labeled datasets. Semi-supervised learning (SSL), although a promising solution to the scarcity of samples, faces challenges such as the imbalance of pseudo-labels and inadequate data representation. In response, this paper presents the Semi-Supervised Adaptive Matrix Machine (SAMM) model, designed for the fault diagnosis of switch machine. SAMM amalgamates semi-supervised learning with adaptive technologies, leveraging adaptive low-rank regularizer to discern the fundamental links between the rows and columns of matrix data and applying adaptive penalty items to correct imbalances across sample categories. This model methodically enlarges its labeled dataset using probabilistic outputs and semi-supervised, automatically adjusting parameters to accommodate diverse data distributions and structural nuances. The SAMM model's optimization process employs the alternating direction method of multipliers (ADMM) to identify solutions efficiently. Experimental evidence from a dataset containing current signals from switch machines indicates that SAMM outperforms existing baseline models, demonstrating its exceptional status diagnostic capabilities in situations where labeled samples are scarce. Consequently, SAMM offers an innovative and effective approach to semi-supervised classification tasks involving matrix data. [ABSTRACT FROM AUTHOR]

Inferring cell type proportions from bulk transcriptome data is crucial in immunology and oncology. Here, we introduce guided LDA deconvolution (GLDADec), a bulk deconvolution method that guides topics using cell type-specific marker gene names to estimate topic distributions for each sample. Through benchmarking using blood-derived datasets, we demonstrate its high estimation performance and robustness. Moreover, we apply GLDADec to heterogeneous tissue bulk data and perform comprehensive cell type analysis in a data-driven manner. We show that GLDADec outperforms existing methods in estimation performance and evaluate its biological interpretability by examining enrichment of biological processes for topics. Finally, we apply GLDADec to The Cancer Genome Atlas tumor samples, enabling subtype stratification and survival analysis based on estimated cell type proportions, thus proving its practical utility in clinical settings. This approach, utilizing marker gene names as partial prior information, can be applied to various scenarios for bulk data deconvolution. GLDADec is available as an open-source Python package at https://github.com/mizuno-group/GLDADec. [ABSTRACT FROM AUTHOR]

Hand gesture recognition is pivotal in facilitating human–machine interaction within the Internet of Things. Nevertheless, it encounters challenges, including labeling expenses and robustness. To tackle these issues, we propose a semi-supervised learning framework guided by pseudo-label consistency. This framework utilizes a dual-branch structure with a mean-teacher network. Within this setup, a global and locally guided self-supervised learning encoder acts as a feature extractor in a teacher–student network to efficiently extract features, maximizing data utilization to enhance feature representation. Additionally, we introduce a pseudo-label Consistency-Guided Mean-Teacher model, where simulated noise is incorporated to generate newly unlabeled samples for the teacher model before advancing to the subsequent stage. By enforcing consistency constraints between the outputs of the teacher and student models, we alleviate accuracy degradation resulting from individual differences and interference from other body parts, thereby bolstering the network's robustness. Ultimately, the teacher model undergoes refinement through exponential moving averages to achieve stable weights. We evaluate our semi-supervised method on two publicly available hand gesture datasets and compare it with several state-of-the-art fully-supervised algorithms. The results demonstrate the robustness of our method, achieving an accuracy rate exceeding 99% across both datasets. [ABSTRACT FROM AUTHOR]

Purpose: Segmentation of orbital tumors in CT images is of great significance for orbital tumor diagnosis, which is one of the most prevalent diseases of the eye. However, the large variety of tumor sizes and shapes makes the segmentation task very challenging, especially when the available annotation data is limited. Methods: To this end, in this paper, we propose a multi‐scale consistent self‐training network (MSCINet) for semi‐supervised orbital tumor segmentation. Specifically, we exploit the semantic‐invariance features by enforcing the consistency between the predictions of different scales of the same image to make the model more robust to size variation. Moreover, we incorporate a new self‐training strategy, which adopts iterative training with an uncertainty filtering mechanism to filter the pseudo‐labels generated by the model, to eliminate the accumulation of pseudo‐label error predictions and increase the generalization of the model. Results: For evaluation, we have built two datasets, the orbital tumor binary segmentation dataset (Orbtum‐B) and the orbital multi‐organ segmentation dataset (Orbtum‐M). Experimental results on these two datasets show that our proposed method can both achieve state‐of‐the‐art performance. In our datasets, there are a total of 55 patients containing 602 2D images. Conclusion: In this paper, we develop a new semi‐supervised segmentation method for orbital tumors, which is designed for the characteristics of orbital tumors and exhibits excellent performance compared to previous semi‐supervised algorithms. [ABSTRACT FROM AUTHOR]

The mean teacher model and its variants, as important methods in semi-supervised learning, have demonstrated promising performance in magnetic resonance imaging (MRI) data segmentation. However, the superior performance of teacher model through exponential moving average (EMA) is limited by the unreliability of unlabeled image, resulting in potentially unreliable predictions. In this paper, we propose a framework to optimized the teacher model with reliable expert-annotated data while preserving the advantages of EMA. To avoid the tight coupling that results from EMA, we leverage data augmentations to provide two distinct perspectives for the teacher and student models. The teacher model adopts weak data augmentation to provide supervision for the student model and optimizes itself with real annotations, while the student uses strong data augmentation to avoid overfitting on noise information. In addition, double softmax helps the model resist noise and continue learning meaningful information from the images, which is a key component in the proposed model. Extensive experiments show that the proposed method exhibits competitive performance on the Left Atrium segmentation MRI dataset (LA) and the Brain Tumor Segmentation MRI dataset (BraTS2019). For the LA dataset, we achieved a dice of 91.02% using only 20% labeled data, which is close to the dice of 91.14% obtained by the supervised approach using 100% labeled data. For the BraTs2019 dataset, the proposed method achieved 1.02% and 1.92% improvement on 5% and 10% labeled data, respectively, compared to the best baseline method on this dataset. This study demonstrates that the proposed model can be a potential candidate for medical image segmentation in semi-supervised learning scenario. [ABSTRACT FROM AUTHOR]