Large Amount Of Unlabeled Data Research Articles

MmGesture: Semi-supervised gesture recognition system using mmWave radar

Gesture recognition has found versatile applications in natural human–computer interaction (HCI). Compared with traditional camera-based or wearable sensors-based solutions, gesture recognition using the millimeter wave (mmWave) radar has attracted growing attention for its characteristics of contact-free, privacy-preserving and less environment-dependence. Recently, most of studies adopted one of the Range Doppler Image (RDI), Range Angle Image (RAI), Doppler Angle Image (DAI) or Micro-Doppler Spectrogram extracted from the raw radar signal as the input of a deep neural network to realize gesture recognition. However, the effectiveness of these four inputs in gesture recognition has attracted little attention so far. Moreover, the lack of large amounts of labeled data restricts the performance of traditional supervised learning network. In this paper, we first conducted extensive experiments to compare the effectiveness of these four inputs in the gesture recognition, respectively. Then we proposed a semi-supervised leaning framework by utilizing few labeled data in the source domain and large amounts of unlabeled data in the target domain. Specially, we combine the ∏-model and some specific data augmentation tricks on the mmWave signal to realize the domain-independent gesture recognition. Extensive experiments on a public mmWave gesture dataset demonstrate the superior effectiveness of the proposed system.

A Cotraining-Based Semisupervised Approach for Remaining-Useful-Life Prediction of Bearings.

The failure of bearings can have a significant negative impact on the safe operation of equipment. Recently, deep learning has become one of the focuses of RUL prediction due to its potent scalability and nonlinear fitting ability. The supervised learning process in deep learning requires a significant quantity of labeled data, but data labeling can be expensive and time-consuming. Cotraining is a semisupervised learning method that reduces the quantity of required labeled data through exploiting available unlabeled data in supervised learning to boost accuracy. This paper innovatively proposes a cotraining-based approach for RUL prediction. A CNN and an LSTM were cotrained on large amounts of unlabeled data to obtain a health indicator (HI), then the monitoring data were entered into the HI and the RUL prediction was realized. The effectiveness of the proposed approach was compared and analyzed against individual CNN and LSTM and the stacking networks SAE+LSTM and CNN+LSTM in the existing literature using RMSE and MAPE values on a PHM 2012 dataset. The results demonstrate that the RMSE and MAPE value of the proposed approach are superior to individual CNN and LSTM, and the RMSE value of the proposed approach is 54.72, which is significantly lower than SAE+LSTM (137.12), and close to CNN+LSTM (49.36). The proposed approach has also been tested successfully on a real-world task and thus has strong application value.

Malware Traffic Classification Using Domain Adaptation and Ladder Network for Secure Industrial Internet of Things

Malware traffic classification (MTC) is a key technology for anomaly and intrusion detection in secure Industrial Internet of Things (IIoT). Traditional MTC methods based on port, payload, and statistic depend on the manual-designed features, which have low accuracy. Recently, deep-learning methods have attracted a significant attention due to their high accuracy in terms of classification. However, in practical application scenarios, deep-learning methods require a large amount of labeled samples for training, while the available labeled samples for training are very rare. Furthermore, the preparation of a large amount of labeled samples requires a lot of labor costs. To solve these problems, this article proposes three methods based on semisupervised learning (SSL), transfer learning (TL), and domain adaptive (DA), respectively. Our proposed methods use a large amount of unlabeled data collected in the Internet traffic, which can greatly improve the classification accuracy with few labeled samples. Then, we use the DA method to solve the mismatch problem between the source domain and the target domain in the TL process. The proposed method is not only applicable to the shallow network but also to the deep neural network structure, and can achieve better classification results. Experimental results show that our proposed methods can satisfy the requirement of MTC in the case of few labeled samples in IIoT. The source code for all the experiments is available at GitHub.The code of this article can be downloaded from GitHub link: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/yzjh/Keras-MTC-DA-Ladder</uri> .

Semi‐supervised pedestrian and face detection via multiple teachers

Semi-supervised learning aims to boost the model performance by large amounts of unlabelled data, thereby reducing the overheads of labelling. For joint pedestrian and face detection in real-world scenarios, the existing semi-supervised object detection methods rarely focus on the category relevance between samples, resulting in unsatisfied classification performance. And it is not effective for existing semi-supervised methods to integrate the categories from two datasets to obtain an ensemble network. In this work, a novel approach aiming to fully utilise category-relevant information is proposed. Firstly, the multi-teacher distillation for decoupling pedestrian and face categories are introduced to eliminate category unfairness in distillation process. Second, a coupled attention module embedded in classification head of the student network is proposed to better grasp the relevance of different categories from teachers and guide distillation. Moreover, the constraint loss is designed for stabilising the training process and better converging, so as to tailor a versatile student. The experimental results on the CrowdHuman and WiderFace benchmarks demonstrate the superiority of the approach over the state-of-the-art methods.

Specific Emitter Identification Based on Self-Supervised Contrast Learning

The current deep learning (DL)-based Specific Emitter Identification (SEI) methods rely heavily on the training of massive labeled data during the training process. However, the lack of labeled data in a real application would lead to a decrease in the method’s identification performance. In this paper, we propose a self-supervised method via contrast learning (SSCL), which is used to extract fingerprint features from unlabeled data. The proposed method uses large amounts of unlabeled data to constitute positive and negative pairs by designing a composition of data augmentation operations for emitter signals. Then, the pairs would be input into the neural network (NN) for feature extraction, and a contrastive loss function is introduced to drive the network to measure the similarity among data. Finally, the identification model can be completed by fixing the parameters of the feature extraction network and fine-tuning with few labeled data. The simulation experiment result shows that, after being fine-tuned, the proposed method can effectively extract fingerprint features. When the SNR is 20 dB, the identification accuracy reaches 94.45%, which is better than the current mainstream DL approaches.

A Semi-Supervised Extreme Learning Machine Algorithm Based on the New Weighted Kernel for Machine Smell

At present, machine sense of smell has shown its important role and advantages in many scenarios. The development of machine sense of smell is inseparable from the support of corresponding data and algorithms. However, the process of olfactory data collection is relatively cumbersome, and it is more difficult to collect labeled data. However, in many scenarios, to use a small amount of labeled data to train a good-performing classifier, it is not feasible to rely only on supervised learning algorithms, but semi-supervised learning algorithms can better cope with only a small amount of labeled data and a large amount of unlabeled data. This study combines the new weighted kernel with SKELM and proposes a semi-supervised extreme learning machine algorithm based on the weighted kernel, SELMWK. The experimental results show that the proposed SELMWK algorithm has good classification performance and can solve the semi-supervised gas classification task of the same domain data well on the used dataset.

Classification of the Epileptic Seizure Onset Zone Based on Partial Annotation

Epilepsy is a chronic disorder caused by excessive electrical discharges. Currently, clinical experts identify the seizure onset zone (SOZ) channel through visual judgment based on long-time intracranial electroencephalogram (iEEG), which is a very time-consuming, difficult and experience-based task. Therefore, there is a need for high-accuracy diagnostic aids to reduce the workload of clinical experts. In this article, we propose a method in which, the iEEG is split into the 20-s segment and for each patient, we ask clinical experts to label a part of the data, which is used to train a model and classify the remaining iEEG data. In recent years, machine learning methods have been successfully applied to solve some medical problems. Filtering, entropy and short-time Fourier transform (STFT) are used for extracting features. We compare them to wavelet transform (WT), empirical mode decomposition (EMD) and other traditional methods with the aim of obtaining the best possible discriminating features. Finally, we look for their medical interpretation, which is important for clinical experts. We achieve high-performance results for SOZ and non-SOZ data classification by using the labeled iEEG data and support vector machine (SVM), fully connected neural network (FCNN) and convolutional neural network (CNN) as classification models. In addition, we introduce the positive unlabeled (PU) learning to further reduce the workload of clinical experts. By using PU learning, we can learn a binary classifier with a small amount of labeled data and a large amount of unlabeled data. This can greatly reduce the amount and difficulty of annotation work by clinical experts. All together, we show that using 105 minutes of labeled data we achieve a classification result of 91.46% on average for multiple patients.

Segmentation only uses sparse annotations: Unified weakly and semi-supervised learning in medical images.

Since segmentation labeling is usually time-consuming and annotating medical images requires professional expertise, it is laborious to obtain a large-scale, high-quality annotated segmentation dataset. We propose a novel weakly- and semi-supervised framework named SOUSA (Segmentation Only Uses Sparse Annotations), aiming at learning from a small set of sparse annotated data and a large amount of unlabeled data. The proposed framework contains a teacher model and a student model. The student model is weakly supervised by scribbles and a Geodesic distance map derived from scribbles. Meanwhile, a large amount of unlabeled data with various perturbations are fed to student and teacher models. The consistency of their output predictions is imposed by Mean Square Error (MSE) loss and a carefully designed Multi-angle Projection Reconstruction (MPR) loss. Extensive experiments are conducted to demonstrate the robustness and generalization ability of our proposed method. Results show that our method outperforms weakly- and semi-supervised state-of-the-art methods on multiple datasets. Furthermore, our method achieves a competitive performance with some fully supervised methods with dense annotation when the size of the dataset is limited.

Dual-MGAN: An Efficient Approach for Semi-supervised Outlier Detection with Few Identified Anomalies

Outlier detection is an important task in data mining, and many technologies for it have been explored in various applications. However, owing to the default assumption that outliers are not concentrated, unsupervised outlier detection may not correctly identify group anomalies with higher levels of density. Although high detection rates and optimal parameters can usually be achieved by using supervised outlier detection, obtaining a sufficient number of correct labels is a time-consuming task. To solve these problems, we focus on semi-supervised outlier detection with few identified anomalies and a large amount of unlabeled data. The task of semi-supervised outlier detection is first decomposed into the detection of discrete anomalies and that of partially identified group anomalies, and a distribution construction sub-module and a data augmentation sub-module are then proposed to identify them, respectively. In this way, the dual multiple generative adversarial networks (Dual-MGAN) that combine the two sub-modules can identify discrete as well as partially identified group anomalies. In addition, in view of the difficulty of determining the stop node of training, two evaluation indicators are introduced to evaluate the training status of the sub-GANs. Extensive experiments on synthetic and real-world data show that the proposed Dual-MGAN can significantly improve the accuracy of outlier detection, and the proposed evaluation indicators can reflect the training status of the sub-GANs.

Image-Level Uncertainty in Pseudo-Label Selection for Semi-Supervised Segmentation.

Advancements in deep learning techniques have proved useful in biomedical image segmentation. However, the large amount of unlabeled data inherent in biomedical imagery, particularly in digital pathology, creates a semi-supervised learning paradigm. Specifically, because of the time consuming nature of producing pixel-wise annotations and the high cost of having a pathologist dedicate time to labeling, there is a large amount of unlabeled data that we wish to utilize in training segmentation algorithms. Pseudo-labeling is one method to leverage the unlabeled data to increase overall model performance. We adapt a method used for image classification pseudo-labeling to select images for segmentation pseudo-labeling and apply it to 3 digital pathology datasets. To select images for pseudo-labeling, we create and explore different thresholds for confidence and uncertainty on an image level basis. Furthermore, we study the relationship between image-level uncertainty and confidence with model performance. We find that the certainty metrics do not consistently correlate with performance intuitively, and abnormal correlations serve as an indicator of a model's ability to produce pseudo-labels that are useful in training. Clinical relevance - The proposed approach adapts image-level confidence and uncertainty measures for segmentation pseudo-labeling on digital pathology datasets. Increased model performance enables better disease quantification for histopathology.

Active Sampling for Text Classification with Subinstance Level Queries

Active learning algorithms are effective in identifying the salient and exemplar samples from large amounts of unlabeled data. This tremendously reduces the human annotation effort in inducing a machine learning model as only a few samples, which are identified by the algorithm, need to be labeled manually. In problem domains like text mining and video classification, human oracles peruse the data instances incrementally to derive an opinion about their class labels (such as reading a movie review progressively to assess its sentiment). In such applications, it is not necessary for the human oracles to review an unlabeled sample end-to-end in order to provide a label; it may be more efficient to identify an optimal subinstance size (percentage of the sample from the start) for each unlabeled sample, and request the human annotator to label the sample by analyzing only the subinstance, instead of the whole data sample. In this paper, we propose a novel framework to address this challenging problem, in an effort to further reduce the labeling burden on the human oracles and utilize the available labeling budget more efficiently. We pose the sample and subinstance size selection as a constrained optimization problem and derive a linear programming relaxation to select a batch of exemplar samples, together with the optimal subinstance size of each, which can potentially augment maximal information to the underlying classification model. Our extensive empirical studies on six challenging datasets from the text mining domain corroborate the practical usefulness of our framework over competing baselines.

Research on Virtual Interactive Animation Design System Based on Deep Learning.

With the rapid development of computer network technology, the advantages of virtual reality technology in the field of instant messaging are becoming more and more significant. Virtual reality technology plays an important role in communication networks, including enhanced resource utilization, device redundancy, immersion, interactivity, conceptualization, and holography. In this paper, we use the basic theory of Restricted Boltzmann Machine to establish a semisupervised spatio-temporal feature model through the animation capture data style recognition problem. The bottom layer can be pretrained with a large amount of unlabeled data to enhance the model's feature perception capability of animation data, and then train the high-level supervised model with the labeled data to finally obtain the model parameters that can be used for the recognition task. The layer-by-layer training method makes the model have good parallelism, that is, when the layer-by-layer training method makes the model well parallelized, that is, when the bottom features cannot effectively represent the animation features, such as overfitting or underfitting, only the bottom model needs to be retrained, while the top model parameters can be kept unchanged. Simulation experiments show that the design assistance time of this paper's scheme for animation is reduced by 10 minutes compared to baseline.

Revisiting instance search: A new benchmark using cycle self-training

Instance search aims at retrieving a particular object instance from a set of scene images. Although studied in previous competitions like TRECVID, there have been limited literature or datasets on this topic. In this paper, to overcome the generalization issue when arbitrary categories are involved in search and to benefit from the large amount of unlabeled data, we propose a cycle self-training framework which trains the instance search pipeline with automatic supervision. Given the two-stage pipeline with a localization and ranking module, the cycle self-training includes a ranker-guided localizer, and a localizer-guided ranker, each carefully designed to handle noisy labels that come with self-supervision. Furthermore, we build and release large-scale groundtruth annotations for instances to facilitate the algorithm evaluation and analysis in this research topic, especially for small objects in complex background. The datasets are publicly available athttps://github.com/instance-search/instance-search. Extensive experiments show the effectiveness of the proposed cycle self-training framework and the superior performance compared with other state-of-the-art methods.

Mutual match for semi-supervised online evolutive learning

Semi-supervised learning (SSL) can utilize a large amount of unlabeled data for self-training and continuous evolution with only a few annotations. This feature makes SSL a potential candidate for dealing with data from changing and real-time environments, where deep-learning models need to be adapting to evolving and nonstable (non-i.i.d.) data streams from the real world, i.e., online evolutive scenarios. However, state-of-the-art SSL methods often have complex model design mechanisms and may cause performance degradation in a generalized and open environment. In an edge computing setup, e.g., typical in modern Internet of Things (IoT) applications, a multi-agent SSL architecture can help resolve generalization problems by sharing knowledge between models. In this paper, we introduce Mutual Match (MM), an online-evolutive SSL algorithm that integrates mutual interactive learning and soft-supervision consistency regularization, as well as unsupervised sample mining. By leveraging extra knowledge in the training process and the interactive collaboration between models, MM surpasses multiple top SSL algorithms in accuracy and convergence efficiency under the same online-evolutive experiment setup. MM simplifies the complexity of model design and follows a unified and easy-to-expandable pipeline, which can be beneficial to tasks with insufficient labeled data and frequently changing data distribution.

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.

Semi-supervised segmentation of echocardiography videos via noise-resilient spatiotemporal semantic calibration and fusion.

We present a novel model for left ventricle endocardium segmentation from echocardiography video, which is of great significance in clinical practice and yet a challenging task due to (1) the severe speckle noise in echocardiography videos, (2) the irregular motion of pathological heart, and (3) the limited training data caused by high annotation cost. The proposed model has three compelling characteristics. First, we propose a novel adaptive spatiotemporal semantic calibration method to align the feature maps of consecutive frames, where the spatiotemporal correspondences are figured out based on feature maps instead of pixels, thereby mitigating the adverse effects of speckle noise in the calibration. Second, we further learn the importance of each feature map of neighbouring frames to the current frame from the temporal perspective so as to distinctively rather than uniformly harness the temporal information to tackle the irregular and anisotropic motions. Third, we integrate these techniques into the mean teacher semi-supervised architecture to leverage a large amount of unlabeled data to improve the segmentation accuracy. We extensively evaluate the proposed method on two public echocardiography video datasets (EchoNet-Dynamic and CAMUS), where the average dice coefficient on the left ventricular endocardium segmentation achieves 92.87% and 93.79%, respectively. Comparisons with state-of-the-art methods also demonstrate the effectiveness of the proposed method by achieving a better segmentation performance with a faster speed.

Semi-supervised pedestrian re-identification via a teacher–student model with similarity-preserving generative adversarial networks

This paper describes a pedestrian re-identification algorithm, which was developed by integrating semi-supervised learning and similarity-preserving generative adversarial networks (SPGAN). The pedestrian re-identification task aimed to rapidly capture the same target using different cameras. Importantly, this process can be applied in the field of security. Because real-life environments are complex, the number of detected identities is uncertain, and the cost of manual labeling is high; therefore, it is difficult to apply the re-identification model based on supervised learning in real-life scenarios. To use the existing labeled dataset and a large amount of unlabeled data in the application environment, this report proposes a semi-supervised pedestrian re-identification model, which combines a teacher–student model with SPGAN. SPGAN was used to reduce the difference between the target domain and the source domain by transferring the style of the labeled dataset from the source domain. Additionally, the dataset from the source domain was used after the style transfer to pre-train the model; this enabled the model to adapt more rapidly to the target domain. The teacher–student model and the transformer model were then employed to generate soft pseudo-labels and hard pseudo-labels (via iterative training) and to update the parameters through distillation learning. Thus, it retained the learned features while adapting to the target domain. Experimental results indicated that the maps of the applied method on the Market-to-Duke, Duke-to-Market, Market-to-MSMT, and Duke-to-MSMT domains were 70.2, 79.3, 30.2, and 33.4, respectively.

Representation learning with deep sparse auto-encoder for multi-task learning

We demonstrate an effective framework to achieve a better performance based on Deep Sparse auto-encoder for Multi-task Learning, called DSML for short. To learn the reconstructed and higher-level features on cross-domain instances for multiple tasks, we combine the labeled and unlabeled data from all tasks to reconstruct the feature representations. Furthermore, we propose the model of Stacked Reconstruction Independence Component Analysis (SRICA for short) for the optimization of feature representations with a large amount of unlabeled data, which can effectively address the redundancy of image data. Our proposed SRICA model is developed from RICA and is based on deep sparse auto-encoder. In addition, we adopt a Semi-Supervised Learning method (SSL for short) based on model parameter regularization to build a unified model for multi-task learning. There are several advantages in our proposed framework as follows: 1) The proposed SRICA makes full use of a large amount of unlabeled data from all tasks. It is used to pursue an optimal sparsity feature representation, which can overcome the over-fitting problem effectively. 2) The deep architecture used in our SRICA model is applied for higher-level and better representation learning, which is designed to train on patches for sphering the input data. 3) Training parameters in our proposed framework has lower computational cost compared to other common deep learning methods such as stacked denoising auto-encoders. Extensive experiments on several real image datasets demonstrate our proposed framework outperforms the state-of-the-art methods.

Efficient Evaluation of Prediction Rules in Semi-Supervised Settings under Stratified Sampling.

In many contemporary applications, large amounts of unlabeled data are readily available while labeled examples are limited. There has been substantial interest in semi-supervised learning (SSL) which aims to leverage unlabeled data to improve estimation or prediction. However, current SSL literature focuses primarily on settings where labeled data is selected uniformly at random from the population of interest. Stratified sampling, while posing additional analytical challenges, is highly applicable to many real world problems. Moreover, no SSL methods currently exist for estimating the prediction performance of a fitted model when the labeled data is not selected uniformly at random. In this paper, we propose a two-step SSL procedure for evaluating a prediction rule derived from a working binary regression model based on the Brier score and overall misclassification rate under stratified sampling. In step I, we impute the missing labels via weighted regression with nonlinear basis functions to account for stratified sampling and to improve efficiency. In step II, we augment the initial imputations to ensure the consistency of the resulting estimators regardless of the specification of the prediction model or the imputation model. The final estimator is then obtained with the augmented imputations. We provide asymptotic theory and numerical studies illustrating that our proposals outperform their supervised counterparts in terms of efficiency gain. Our methods are motivated by electronic health record (EHR) research and validated with a real data analysis of an EHR-based study of diabetic neuropathy.

Fully used reliable data and attention consistency for semi-supervised learning

Large labeled datasets represent human labor’s costly consumption of resources. Therefore, semi-supervised learning leverages a large amount of unlabeled data to improve the training results in limited labels. Many methods of semi-supervised learning utilize diverse data augmentations to improve model learning and the classification rule from these changes, requiring models to spend a lot of time to adapt to the changes. Besides, reducing the noise in trained unlabeled data is also an issue that is often discussed in semi-supervised learning so that the inference from error predictions can be reduced. It may define that the data, of which the probability predicted from the model is higher than a threshold, as confident and then only train on those high-confidence unlabeled data so that the model avoids the influence from deviation of the error caused by unlabeled data predictions. However, it also leads to the fact that many unlabeled data cannot be effectively used. Thus, this study proposes a semi-supervised framework, including Attention Consistency (AC) and One Supervised (OS) algorithms, which improves efficiency and performance of the model learning by guiding the model to pay attention to classified features and judging whether the model cannot be effectively trained in existing reliable data. This way, the model fully uses unlabeled data to train. The experiment results and comparisons show that similar results can be reached using other methods within a shorter training process. This paper also analyzes the distribution of feature results and proposes a new measurement to find out distribution information.