Unlabeled Information Research Articles

Efficient semiparametric estimation in two‐sample comparison via semisupervised learning

AbstractWe develop a general semisupervised framework for statistical inference in the two‐sample comparison setting. Although the supervised Mann–Whitney statistic outperforms many estimators in the two‐sample problem for nonnormally distributed responses, it is excessively inefficient because it ignores large amounts of unlabelled information. To borrow strength from unlabelled data, we propose a class of efficient and adaptive estimators that use two‐step semiparametric imputation. The probabilistic index model is adopted primarily to achieve dimension reduction for multivariate covariates, and a follow‐up reweighting step balances the contributions of labelled and unlabelled data. The asymptotic properties of our estimator are derived with variance comparison through a phase diagram. Efficiency theory shows our estimators achieve the semiparametric variance lower bound if the probabilistic index model is correctly specified, and are more efficient than their supervised counterpart when the model is not degenerate. The asymptotic variance is estimated through a two‐step perturbation resampling procedure. To gauge the finite sample performance, we conducted extensive simulation studies which verify the adaptive nature of our methods with respect to model misspecification. To illustrate the merits of our proposed method, we analyze a dataset concerning homelessness in Los Angeles.

TopoGCL: Topological Graph Contrastive Learning

Graph contrastive learning (GCL) has recently emerged as a new concept which allows for capitalizing on the strengths of graph neural networks (GNNs) to learn rich representations in a wide variety of applications which involve abundant unlabeled information. However, existing GCL approaches largely tend to overlook the important latent information on higher-order graph substructures. We address this limitation by introducing the concepts of topological invariance and extended persistence on graphs to GCL. In particular, we propose a new contrastive mode which targets topological representations of the two augmented views from the same graph, yielded by extracting latent shape properties of the graph at multiple resolutions. Along with the extended topological layer, we introduce a new extended persistence summary, namely, extended persistence landscapes (EPL) and derive its theoretical stability guarantees. Our extensive numerical results on biological, chemical, and social interaction graphs show that the new Topological Graph Contrastive Learning (TopoGCL) model delivers significant performance gains in unsupervised graph classification for 8 out of 12 considered datasets and also exhibits robustness under noisy scenarios.

A semi-supervised approach for the integration of multi-omics data based on transformer multi-head self-attention mechanism and graph convolutional networks.

Comprehensive analysis of multi-omics data is crucial for accurately formulating effective treatment plans for complex diseases. Supervised ensemble methods have gained popularity in recent years for multi-omics data analysis. However, existing research based on supervised learning algorithms often fails to fully harness the information from unlabeled nodes and overlooks the latent features within and among different omics, as well as the various associations among features. Here, we present a novel multi-omics integrative method MOSEGCN, based on the Transformer multi-head self-attention mechanism and Graph Convolutional Networks(GCN), withtheaimofenhancingtheaccuracyofcomplexdiseaseclassification. MOSEGCN first employs the Transformer multi-head self-attention mechanism and Similarity Network Fusion (SNF) to separately learn the inherent correlations of latent features within and among different omics, constructing a comprehensive view of diseases. Subsequently, it feeds the learned crucial information into a self-ensembling Graph Convolutional Network (SEGCN) built upon semi-supervised learning methods for training and testing, facilitating a better analysis and utilization of information from multi-omics data to achieve precise classification of disease subtypes. The experimental results show that MOSEGCN outperforms several state-of-the-art multi-omics integrative analysis approaches on three types of omics data: mRNA expression data, microRNA expression data, and DNA methylation data, with accuracy rates of 83.0% for Alzheimer's disease and 86.7% for breast cancer subtyping. Furthermore, MOSEGCN exhibits strong generalizability on the GBM dataset, enabling the identification of important biomarkers for related diseases. MOSEGCN explores the significant relationship information among different omics and within each omics' latent features, effectively leveraging labeled and unlabeled information to further enhance the accuracy of complex disease classification. It also provides a promising approach for identifying reliable biomarkers, paving the way for personalized medicine.

A constrastive semi-supervised deep learning framework for land cover classification of satellite time series with limited labels

In this work, we present a new semi-supervised learning framework to cope with satellite image time series (SITS) classification in a data paucity scenario, considering extreme low levels of supervision. The proposed methodology, referred as S3ITS (Semi-Supervised Satellite Image Time Series classification method), is based on temporal convolutional neural networks and it takes advantage of both labelled and unlabelled information. S3ITS enforces the data to be projected in a discriminative manifold via contrastive learning, in order to produce a data representation where samples belonging to the same category are closer than the ones belonging to different ones. Pseudo-labelling is employed on unlabelled samples to take the most out of the available unlabelled information. Experiments on two study sites described by SITS of Sentinel-2 images highlight the quality of the proposed method with respect to common SITS-based classification methods and recent machine learning approaches especially tailored for the semi-supervised classification of multi-variate time series data.

Self-Adaptive Imbalanced Domain Adaptation With Deep Sparse Autoencoder

Domain adaptation aims to transfer knowledge between different domains to develop an effective hypothesis in the target domain with scarce labeled data, which is an effective method for remedying the problem of labeled data requirement in deep learning. In reality, it is unavoidable that the dataset has a large gap in the number of positive and negative instances across different categories in source and target domains, which is the imbalanced domain adaptation problem. However, since the imbalanced degree always varies greatly in different source and target domains datasets, most existing imbalanced domain adaptation models fix the imbalanced parameters which can not adapt to the change of the proportion between positive and negative instances in different domains. To address this problem, in this paper, we propose a self-adaptive imbalanced domain adaptation method via deep sparse autoencoder, which can adjust the model automatically according to the imbalanced extent for bridging the chasm of domains. More specifically, the self-adaptive imbalanced cross entropy loss is designed for emphasizing more on minority categories and compensating the bias of training loss automatically. In addition, to alleviate the deficient problem of labeled data, we further propose the unlabeled information incorporating method by minimizing the distribution discrepancy of high-level representation space between the source and target domains. Experiments on several real-world datasets demonstrate the effectiveness of our method compared to other state-of-the-art methods.

Using Unlabeled Information of Embryo Siblings from the Same Cohort Cycle to Enhance In Vitro Fertilization Implantation Prediction.

High-content time-lapse embryo imaging assessed by machine learning is revolutionizing the field of in vitro fertilization (IVF). However, the vast majority of IVF embryos are not transferred to the uterus, and these masses of embryos with unknown implantation outcomes are ignored in current efforts that aim to predict implantation. Here, whether, and to what extent the information encoded within "sibling" embryos from the same IVF cohort contributes to the performance of machine learning-based implantation prediction is explored. First, it is shown that the implantation outcome is correlated with attributes derived from the cohort siblings. Second, it is demonstrated that this unlabeled data boosts implantation prediction performance. Third, the cohort properties driving embryo prediction, especially those that rescued erroneous predictions, are characterized. The results suggest that predictive models for embryo implantation can benefit from the overlooked, widely available unlabeled data of sibling embryos by reducing the inherent noise of the individual transferred embryo.

A railway intrusion detection method based on decomposition and semi-supervised learning for accident protection

In recent years, video surveillance has become increasingly popular in railway intrusion detection. However, it is still quite challenging to detect the intruded objects efficiently and accurately because: (a) The backgrounds of video frames generated by the fixed cameras are similar and only few intrusive frames are available, resulting in a lack of diversity among video frames, and further leading to over fitting of the detection models during training; (b) The intrusion of small targets or targets far from the location of camera exhibits sparsity relative to the wide monitoring view of the camera, which challenges the detection of such targets in a complex background; (c) The extreme imbalance between non-intrusive frames and intrusive frames, as well as a large number of unlabeled frames, hinder the effective training of the detection model and weaken its capacity of generalization. To tackle the above issue, this article develops an effective intrusion detection method by combining low-rank and sparse decomposition (LRSD) and Semi-supervised Support Vector Domain Description (Semi-SVDD). Firstly, LRSD is used to decompose the monitored video into a background and a foreground. Then, based on the semantic segmentation method, we extract the mask of the track region in the decomposed background, which is used to mask the foreground. Next, by using both the labeled and unlabeled frames of the masked foreground, Semi-SVDD is established for the intrusion detection. Numerical results show that the removal of background interference and the combination of the labeled and unlabeled information help to improve the performance of the proposed method, and thus superior to benchmark methods.

CVCL: Context-aware Voxel-wise Contrastive Learning for label-efficient multi-organ segmentation

Despite the significant performance improvement on multi-organ segmentation with supervised deep learning-based methods, the label-hungry nature hinders their applications in practical disease diagnosis and treatment planning. Due to the challenges in obtaining expert-level accurate, densely annotated multi-organ datasets, label-efficient segmentation, such as partially supervised segmentation trained on partially labeled datasets or semi-supervised medical image segmentation, has attracted increasing attention recently. However, most of these methods suffer from the limitation that they neglect or underestimate the challenging unlabeled regions during model training. To this end, we propose a novel Context-aware Voxel-wise Contrastive Learning method, referred as CVCL, to take full advantage of both labeled and unlabeled information in label-scarce datasets for a performance improvement on multi-organ segmentation. Experimental results demonstrate that our proposed method achieves superior performance than other state-of-the-art methods.

A collaborative central domain adaptation approach with multi-order graph embedding for bearing fault diagnosis under few-shot samples

Effective fault diagnosis is a prerequisite for ensuring the safe, stable and long-term operation of many rotating machinery. With the rapid development of measurement, sensor and computing technologies, measurement data presents a high-dimensional and massive distribution. This makes the valuable fault information in samples sparse. Moreover, industrial data can only present the distribution state of few-shot unlabeled information. In addition, the vibration signal of bearing faults contains noise interference, leading to poor stability and low efficiency of most models. In this study, we propose an approach for rolling bearing faults diagnosis under few-shot samples. It consists of a multi-order graph embedding stacked denoising auto encoder optimized by an improved sine–cosine​ algorithm (MGE-ISCA-SDAE) and a collaborative central domain adaptation (CCDA). First, a multi-order graph embedding model and an ISCA-based strategy are designed to improve the SDAE, thereby improving the feature extraction effect. To overcome the sparseness of valuable information, we design a CCDA model that learns the fault features using the labeled samples. Subsequently, it is transferred to the target domain of few-shot labeled samples for adaptation. Finally, the intelligent diagnosis is achieved under few-shot samples. We conduct experiments with four datasets. The results show that the MGE-ISCA-SDAE can extract the time–frequency high-level fault features. The CCDA model can transfer the fault samples well. When there are fewer fault samples, our approach has outstanding advantages.

CPAE: Contrastive predictive autoencoder for unsupervised pre-training in health status prediction

Background and objectiveFully-supervised learning approaches have shown promising results in some health status prediction tasks using Electronic Health Records (EHRs). These traditional approaches rely on sufficient labeled data to learn from. However, in practice, acquiring large-scaled labeled medical data for various prediction tasks is often not feasible. Thus, it is of great interest to utilize contrastive pre-training to leverage the unlabeled information. MethodsIn this work, we propose a novel data-efficient framework, contrastive predictive autoencoder (CPAE), to first learn without labels from the EHR data in the pre-training process, and then fine-tune on the downstream tasks. Our framework comprises of two parts: (i) a contrastive learning process, inherited from contrastive predictive coding (CPC), which aims to extract global slow-varying features, and (ii) a reconstruction process, which forces the encoder to capture local features. We also introduce the attention mechanism in one variant of our framework to balance the above two processes. ResultsExperiments on real-world EHR dataset verify the effectiveness of our proposed framework on two downstream tasks (i.e., in-hospital mortality prediction and length-of-stay prediction), compared to their supervised counterparts, the CPC model, and other baseline models. ConclusionsBy comprising of both contrastive learning components and reconstruction components, CPAE aims to extract both global slow-varying information and local transient information. The best results on two downstream tasks are all achieved by CPAE. The variant AtCPAE is particularly superior when fine-tuned on very small training data. Further work may incorporate techniques of multi-task learning to optimize the pre-training process of CPAEs. Moreover, this work is based on the benchmark MIMIC-III dataset which only includes 17 variables. Future work may extend to a larger number of variables.

Two-stage natural scene image classification with noise discovering and label-correlation mining

In natural scene classification, it is common that one natural scene image may belong to multiple categories concurrently; as a result, multi-label learning has become a research hotspot. Despite recent rapid developments in multi-label learning, the increasing amount of high-dimensional data poses great challenges—such as redundant features and high computational costs—to conventional multi-label learning models. Most contemporary strategies for dealing with this issue depend on forcing feature learning into multi-label models. Notably, however, these approaches seldom pay attention to the label correlation and propagation in the feature subspace. To address this issue, we introduce an alternative multi-label feature learning solution that incorporates both labeled and unlabeled information. Unlike existing multi-label learning models , which rely on clean and trustworthy training datasets, we argue that in semi-supervised learning scenarios, the unlabeled data can be easily corrupted by noise or outliers, which causes the model performance to degrade. We next extract the label correlation and propagate the label information to discover the noise or outliers. Subsequently, our model adaptively searches the optimal feature subspace to reduce the influence of redundant features for high-dimensional data. The effectiveness of the proposed model is demonstrated by experimental observations on artificial and real-world datasets.

Co-Learning Bayesian Optimization.

Bayesian optimization (BO) is well known to be sample efficient for solving black-box problems. However, BO algorithms may get stuck in suboptimal solutions even with plenty of samples. Intrinsically, such a suboptimal problem of BO can attribute to the poor surrogate accuracy of the trained Gaussian process (GP), particularly that in the regions where the optimal solutions locate. Hence, we propose to build multiple GP models instead of a single GP surrogate to complement each other, thus resolving the suboptimal problem of BO. Nevertheless, according to the bias-variance tradeoff equation, the individual prediction errors can increase when increasing the diversity of models, which may lead even worse overall surrogate accuracy. On the other hand, based on the theory of the Rademacher complexity, it has been proven that exploiting the agreement of models on unlabeled information can reduce the complexity of hypothesis space, therefore achieving the required surrogate accuracy with fewer samples. Such value of model agreement has been extensively demonstrated for co-training style algorithms to boost model accuracy with a small portion of samples. Inspired by the above, we propose a novel BO algorithm labeled as co-learning BO (CLBO), which exploits both model diversity and agreement on unlabeled information to improve the overall surrogate accuracy with limited samples, therefore achieving more efficient global optimization. Through tests on five numerical toy problems and three engineering benchmarks, the effectiveness of the proposed CLBO has been well demonstrated.

Contrastive Graph Convolutional Networks with adaptive augmentation for text classification

Text classification is an important research topic in natural language processing (NLP), and Graph Neural Networks (GNNs) have recently been applied in this task. However, in existing graph-based models, text graphs constructed by rules are not real graph data and introduce massive noise. More importantly, for fixed corpus-level graph structure, these models cannot sufficiently exploit the labeled and unlabeled information of nodes. Meanwhile, contrastive learning has been developed as an effective method in graph domain to fully utilize the information of nodes. Therefore, we propose a new graph-based model for text classification named CGA2TC, which introduces contrastive learning with an adaptive augmentation strategy into obtaining more robust node representation. First, we explore word co-occurrence and document word relationships to construct a text graph. Then, we design an adaptive augmentation strategy for the text graph with noise to generate two contrastive views that effectively solve the noise problem and preserve essential structure. Specifically, we design noise-based and centrality-based augmentation strategies on the topological structure of text graph to disturb the unimportant connections and thus highlight the relatively important edges. As for the labeled nodes, we take the nodes with same label as multiple positive samples and assign them to anchor node, while we employ consistency training on unlabeled nodes to constrain model predictions. Finally, to reduce the resource consumption of contrastive learning, we adopt a random sample method to select some nodes to calculate contrastive loss. The experimental results on several benchmark datasets can demonstrate the effectiveness of CGA2TC on the text classification task.

Automated Registration of Multiangle SAR Images Using Artificial Intelligence

Traditionally, nonlinear data processing has been approached via the use of polynomial filters, which are straightforward expansions of many linear methods, or through the use of neural network techniques. In contrast to linear approaches, which often provide algorithms that are simple to apply, nonlinear learning machines such as neural networks demand more computing and are more likely to have nonlinear optimization difficulties, which are more difficult to solve. Kernel methods, a recently developed technology, are strong machine learning approaches that have a less complicated architecture and give a straightforward way to transforming nonlinear optimization issues into convex optimization problems. Typical analytical tasks in kernel-based learning include classification, regression, and clustering, all of which are compromised. For image processing applications, a semisupervised deep learning approach, which is driven by a little amount of labeled data and a large amount of unlabeled data, has shown excellent performance in recent years. For their part, today’s semisupervised learning methods operate on the assumption that both labeled and unlabeled information are distributed in a similar manner, and their performance is mostly impacted by the fact that the two data sets are in a similar state of distribution as well. When there is out-of-class data in unlabeled data, the system’s performance will be adversely affected. When used in real-world applications, the capacity to verify that unlabeled data does not include data that belongs to a different category is difficult to obtain, and this is especially true in the field of synthetic aperture radar image identification (SAR). Using threshold filtering, this work addresses the problem of unlabeled input, including out-of-class data, having a detrimental influence on the performance of the model when it is utilized to train the model in a semisupervised learning environment. When the model is being trained, unlabeled data that does not belong to a category is filtered out by the model using two different sets of data that the model selects in order to optimize its performance. A series of experiments was carried out on the MSTAR data set, and the superiority of our method was shown when it was compared against a large number of current semisupervised classification algorithms of the highest level of sophistication. This was especially true when the unlabeled data had a significant proportion of data that did not fall into any of the categories. The performance of each kernel function is tested independently using two metrics, namely, the false alarm (FA) and the target miss (TM), respectively. These factors are used to calculate the proportion of incorrect judgments made using the techniques.

A System of Remote Patients’ Monitoring and Alerting Using the Machine Learning Technique

Machine learning has become an essential tool in daily life, or we can say it is a powerful tool in the majority of areas that we wish to optimize. Machine learning is being used to create techniques that can learn from labelled or unlabeled information, as well as learn from their surroundings. Machine learning is utilized in various areas, but mainly in the healthcare industry, where it provides significant advantages via appropriate decision and prediction methods. The proposed work introduces a remote system that can continuously monitor the patient and can produce an alert whenever necessary. The proposed methodology makes use of different machine learning algorithms along with cloud computing for continuous data storage. Over the years, these technologies have resulted in significant advancements in the healthcare industry. Medical professionals utilize machine learning tools and methods to analyse medical data in order to detect hazards and offer appropriate diagnosis and treatment. The scope of remote healthcare includes anything from tracking chronically sick patients, elderly people, preterm children, and accident victims. The current study explores the machine learning technologies’ capability of monitoring remote patients and alerts their current condition through the remote system. New advances in contactless observation demonstrate that it is only necessary for the patient to be present within a few meters of the sensors for them to work. Sensors connected to the body and environmental sensors connected to the surroundings are examples of the technology available.

Medical Instrument Segmentation in 3D US by Hybrid Constrained Semi-Supervised Learning.

Medical instrument segmentation in 3D ultrasound is essential for image-guided intervention. However, to train a successful deep neural network for instrument segmentation, a large number of labeled images are required, which is expensive and time-consuming to obtain. In this article, we propose a semi-supervised learning (SSL) framework for instrument segmentation in 3D US, which requires much less annotation effort than the existing methods. To achieve the SSL learning, a Dual-UNet is proposed to segment the instrument. The Dual-UNet leverages unlabeled data using a novel hybrid loss function, consisting of uncertainty and contextual constraints. Specifically, the uncertainty constraints leverage the uncertainty estimation of the predictions of the UNet, and therefore improve the unlabeled information for SSL training. In addition, contextual constraints exploit the contextual information of the training images, which are used as the complementary information for voxel-wise uncertainty estimation. Extensive experiments on multiple ex-vivo and in-vivo datasets show that our proposed method achieves Dice score of about 68.6%-69.1% and the inference time of about 1 sec. per volume. These results are better than the state-of-the-art SSL methods and the inference time is comparable to the supervised approaches.

TensorFlow-based smart home using semi-supervised deep learning with context-awareness

In recent years, Smart homes play a significant role in improving the quality of human life due to the rapid proliferation of the Internet of Things (IoT) technology. The previous research works on the smart home system have adopted the machine learning and deep learning algorithms to predict the sequential activities in the smart home. This work presents a model of SMART home automation with Context-Awareness using Stacked AutoEncoder (SAE) -Long Short-Term Memory (LSTM) in TensorFlow (SMART-CAST). The SMART-CAST approach comprises three main processes, including the integration of internal and external home data, SAE-assisted unsupervised learning, and LSTM with back propagation-assisted supervised learning. By inter-linking the spatial and temporal attribute-values, the SMART-CAST unifies the smart home and weather data for facilitating decision-making. It employs the SAE to generate the compressed representation of the unified smart home data from the unlabeled information. In consequence, the SMART-CAST approach applies the LSTM with the extracted compressed representation for learning the labeled data and updates the weight of the LSTM through backpropagation to predict the sequential activities in the smart home system. To further improve the decision-making performance, the experimental model executes the proposed semi-supervised deep learning algorithm in the TensorFlow deep learning framework.

Fish Segmentation in Sonar Images by Mask R-CNN on Feature Maps of Conditional Random Fields.

Imaging sonar systems are widely used for monitoring fish behavior in turbid or low ambient light waters. For analyzing fish behavior in sonar images, fish segmentation is often required. In this paper, Mask R-CNN is adopted for segmenting fish in sonar images. Sonar images acquired from different shallow waters can be quite different in the contrast between fish and the background. That difference can make Mask R-CNN trained on examples collected from one fish farm ineffective to fish segmentation for the other fish farms. In this paper, a preprocessing convolutional neural network (PreCNN) is proposed to provide “standardized” feature maps for Mask R-CNN and to ease applying Mask R-CNN trained for one fish farm to the others. PreCNN aims at decoupling learning of fish instances from learning of fish-cultured environments. PreCNN is a semantic segmentation network and integrated with conditional random fields. PreCNN can utilize successive sonar images and can be trained by semi-supervised learning to make use of unlabeled information. Experimental results have shown that Mask R-CNN on the output of PreCNN is more accurate than Mask R-CNN directly on sonar images. Applying Mask R-CNN plus PreCNN trained for one fish farm to new fish farms is also more effective.

Monocular Depth Estimation using Integrated Model with Multi-task Learning and Multi-stream

Scene understanding is a central problem in a field of computer vision. Depth estimation, in particular, is one of the important applications in scene understanding, robotics, and 3-D reconstruction. Estimating a dense depth map from a single image is receiving increased attention because a monocular camera is popular, small and suitable for a wide range of environments. In addition, both multi-task learning and multi-stream, which use unlabeled information, improve the monocular depth estimation efficiently. However, there are only a few networks optimized for both of them. Therefore, in this paper, we propose a monocular depth estimation task with a multi-task and multistream network architecture. Furthermore, the integrated network which we develop makes use of depth gradient information and can be applied to both supervised and unsupervised learning. In our experiments, we confirmed that our supervised learning architecture improves the accuracy of depth estimation by 0.13 m on average. Additionally, the experimental result on unsupervised learning found that it improved structure-from-motion performance.

PTN: A Poisson Transfer Network for Semi-supervised Few-shot Learning

The predicament in semi-supervised few-shot learning (SSFSL) is to maximize the value of the extra unlabeled data to boost the few-shot learner. In this paper, we propose a Poisson Transfer Network (PTN) to mine the unlabeled information for SSFSL from two aspects. First, the Poisson Merriman–Bence–Osher (MBO) model builds a bridge for the communications between labeled and unlabeled examples. This model serves as a more stable and informative classifier than traditional graph-based SSFSL methods in the message-passing process of the labels. Second, the extra unlabeled samples are employed to transfer the knowledge from base classes to novel classes through contrastive learning. Specifically, we force the augmented positive pairs close while push the negative ones distant. Our contrastive transfer scheme implicitly learns the novel-class embeddings to alleviate the over-fitting problem on the few labeled data. Thus, we can mitigate the degeneration of embedding generality in novel classes. Extensive experiments indicate that PTN outperforms the state-of-the-art few-shot and SSFSL models on miniImageNet and tieredImageNet benchmark datasets.