Semi-supervised One-pass Learning under Distribution Shift

Training deep neural networks for image recognition often requires large-scale human annotated data. To reduce the reliance of deep neural solutions on labeled data, state-of-the-art semi-supervised methods have been proposed in the literature. Nonetheless, the use of such semi-supervised methods has been quite rare in the field of facial expression recognition (FER). In this paper, we present a comprehensive study on recently proposed state-of-the-art semi-supervised learning methods in the context of FER. We conduct comparative study on eight semi-supervised learning methods, namely Pi-Model, Pseudo-label, Mean-Teacher, VAT, MixMatch, ReMixMatch, UDA, and FixMatch, on three FER datasets (FER13, RAF-DB, and AffectNet), when various amounts of labeled samples are used. We also compare the performance of these methods against fully-supervised training. Our study shows that when training existing semi-supervised methods on as little as 250 labeled samples per class can yield comparable performances to that of fully-supervised methods trained on the full labeled datasets. To facilitate further research in this area, we make our code publicly available at: https://github.com/ShuvenduRoy/SSL_FER.

Real-time monitoring microblog data can find sensitive information in time and provide help for public sentiment management and control. However, it needs processing large-scale data stream. MapReduce is a framework of processing large-scale data in batch mode, its purpose is to increase throughput, but its real-time performance is limited. Aiming at the real-time performance limitation of MapReduce, RT-SSP Real-Time Staged Stream Processing, a hybrid staged real-time stream processing scheme both for batch and real-time processing was proposed. By this method large-scale high-speed data stream is locally processed in stages, the communication cost is reduced by storing intermediate results to local node, and key technologies such as cache optimization are used to realize high concurrent read and write. Experiments show that RT-SSP scheme can improve the real-time performance of processing large-scale microblog data stream and achieve speed-up ratio of about 2.3.

Automated diagnosis of infant hip dysplasia is heavily affected by the individual differences among infants and ultrasound machines. Hip sonographic images of 493 infants from various ultrasound machines were collected in the Department of Orthopedics in Yangzhou Maternal and Child Health Care Service Centre. Herein, we propose a semi-supervised learning method based on a feature pyramid network (FPN) and a contrastive learning scheme based on a Siamese architecture. A large amount of unlabeled data of ultrasound images was used via the Siamese network in the pre-training step, and then a small amount of annotated data for anatomical structures was adopted to train the model for landmark identification and standard plane recognition. The method was evaluated on our collected dataset. The method achieved a mean Dice similarity coefficient (DSC) of 0.7873 and a mean Hausdorff distance (HD) of 5.0102 in landmark identification, compared to the model without contrastive learning, which had a mean DSC of 0.7734 and a mean HD of 6.1586. The accuracy, precision, and recall of standard plane recognition were 95.4%, 91.64%, and 94.86%, respectively. The corresponding area under the curve (AUC) was 0.982. This study proposes a semi-supervised deep learning method following Graf's principle, which can better utilize a large volume of ultrasound images from various devices and infants. This method can identify the landmarks of infant hips more accurately than manual operators, thereby improving the efficiency of diagnosis of infant hip dysplasia.

Semi-supervised kernel learning methods have been received much more attention in the past few years. Traditional semi-supervised Non-Parametric Kernel Learning (NPKL) methods usually formulate the learning task as a Semi-Definite Programming (SDP) problem, which is very time consuming. Although some fast semi-supervised NPKL methods have been proposed recently, they usually scale very poorly. Furthermore, many semi-supervised NPKL methods are developed based on the manifold assumption. But, such an assumption might be invalid when handling some high-dimensional and sparse data, which has severely negative effect on the performance of learning algorithms. In this paper, we propose a more efficient semi-supervised NPKL method, which can effectively learn a low-rank kernel matrix from must-link and cannot-link constraints. Specially, by virtue of the nonlinear spectral embedded technique based on extreme learning machine (ELM), the proposed method has the ability of coping with data points that do not have a clear manifold structure in a low dimensional space. The proposed method is formulated as a trace ratio optimization problem, which is combined with dimensionality reduction in ELM feature space and aims to find optimal low-rank kernel matrices. The proposed optimization problem can be solved much more efficiently than SDP solvers. Extensive experiments have validated the superior performance of the proposed method compared to state-of-the-art semi-supervised kernel learning methods.

This paper investigates the use of semi-supervised learning (SSL) for predicting Alzheimers Disease (AD) conversion in Mild Cognitive Impairment (MCI) patients based on Magnetic Resonance Imaging (MRI). SSL methods differ from standard supervised learning methods in that they make use of unlabeled data - in this case data from MCI subjects whose final diagnosis is not yet known. We compare two widely used semi-supervised methods (low density separation (LDS) and semi-supervised discriminant analysis (SDA)) to the corresponding supervised methods using real and synthetic MRI data of MCI subjects. With simulated data, using SSL instead of supervised learning led to higher classification performance in certain cases, however, the applicability of semi-supervised methods depended strongly on the data distributions. With real MRI data, the SSL methods achieved significantly better classification performances over supervised methods. Moreover, even using a small number of unlabeled samples improved the AD conversion predictions.

With the explosive growth of surveillance data, exact match queries become much more difficult for its high dimension and high volume. Owing to its good balance between the retrieval performance and the computational cost, hash learning technique is widely used in solving approximate nearest neighbor search problems. Dimensionality reduction plays a critical role in hash learning, as its target is to preserve the most original information into low-dimensional vectors. However, the existing dimensionality reduction methods neglect to unify diverse resources in original space when learning a downsized subspace. In this article, we propose a numeric and semantic consistency semi-supervised hash learning method, which unifies the numeric features and supervised semantic features into a low-dimensional subspace before hash encoding, and improves a multiple table hash method with complementary numeric local distribution structure. A consistency-based learning method, which confers the meaning of semantic to numeric features in dimensionality reduction, is presented. The experiments are conducted on two public datasets, that is, a web image NUS-WIDE and text dataset DBLP. Experimental results demonstrate that the semi-supervised hash learning method, with the consistency-based information subspace, is more effective in preserving useful information for hash encoding than state-of-the-art methods and achieves high-quality retrieval performance in multi-table context.

In the literature, most existing graph-based semi-supervised learning methods only use the label information of observed samples in the label propagation stage, while ignoring such valuable information when learning the graph. In this paper, we argue that it is beneficial to consider the label information in the graph learning stage. Specifically, by enforcing the weight of edges between labeled samples of different classes to be zero, we explicitly incorporate the label information into the state-of-the-art graph learning methods, such as the low-rank representation (LRR), and propose a novel semi-supervised graph learning method called semi-supervised low-rank representation. This results in a convex optimization problem with linear constraints, which can be solved by the linearized alternating direction method. Though we take LRR as an example, our proposed method is in fact very general and can be applied to any self-representation graph learning methods. Experiment results on both synthetic and real data sets demonstrate that the proposed graph learning method can better capture the global geometric structure of the data, and therefore is more effective for semi-supervised learning tasks.

Online semi-supervised active learning ensemble classification for evolving imbalanced data streams

The article examines the application of semi-supervised learning and computer vision methods for segmentation of spine MR images and diagnosis of deformation of intervertebral discs. Existing neural network architectures for spine MR image segmentation and semi-supervised learning methods used in medical image segmentation are reviewed. A system for diagnosis deformation of intervertebral discs is proposed, which consists of two modules: a segmentation module and a diagnostic module. The implementation of the proposed system using two convolutional neural networks is presented: U-Net for segmentation of spine MR images and ResNet for classification of the degree of deformation of each intervertebral disc based on the Pfirrmann classification of degenerative changes. The software implementation of the medical diagnosis system was developed in the Python programming language using the PyTorch library. Neural networks were trained on an open dataset of spine MR images using a variant of the Mean Teacher semi-supervised learning method. As a result of the verification, it was found that the system is capable of performing segmentation with high accuracy. It was found that the exact prediction of the grade of degenerative changes according to Pfirrmann remains a difficult task, but the introduction of another classification made it possible to increase the accuracy of diagnosis of deformation of intervertebral discs. The proposed medical system involves the addition of new diagnostic modules, which makes it possible to use it for the comprehensive analysis of various spine diseases

Semi-supervised distance metric learning based on local linear regression for data clustering

A novel semi-supervised meta learning method for subject-transfer brain–computer interface

Semi-supervised learning (SSL) is one of the dominant approaches to address the annotation bottleneck of supervised learning. Recent SSL methods can effectively leverage a large repository of unlabeled data to improve performance while relying on a small set of labeled data. One common assumption in most SSL methods is that the labeled and unlabeled data are from the same data distribution. However, this is hardly the case in many real-world scenarios, which limits their applicability. In this work, instead, we attempt to solve the challenging open-world SSL problem that does not make such an assumption. In the open-world SSL problem, the objective is to recognize samples of known classes, and simultaneously detect and cluster samples belonging to novel classes present in unlabeled data. This work introduces OpenLDN that utilizes a pairwise similarity loss to discover novel classes. Using a bi-level optimization rule this pairwise similarity loss exploits the information available in the labeled set to implicitly cluster novel class samples, while simultaneously recognizing samples from known classes. After discovering novel classes, OpenLDN transforms the open-world SSL problem into a standard SSL problem to achieve additional performance gains using existing SSL methods. Our extensive experiments demonstrate that OpenLDN outperforms the current state-of-the-art methods on multiple popular classification benchmarks while providing a better accuracy/training time trade-off. Code: https://github.com/nayeemrizve/OpenLDN.KeywordsOpen-worldSemi-supervised learningNovel classes

Data streams may encounter data distribution changes, which can significantly impair the accuracy of models. Concept drift detection tracks data distribution changes and signals when to update models. Many drift detection methods apply thresholds to distinguish between drift or non-drift streams and to claim their method outperforms others with non-aligned drift thresholds. We consider that selecting a proper drift threshold could be more important than developing a new drift detection algorithm, and different drift detection algorithms may end up with very similar performance with aligned drift thresholds. To better understand this process, we propose a novel threshold selection algorithm to align the drift thresholds of a set of algorithms so that they are all at the same sensitivity level. Based on comprehensive experiment evaluations, we observed that several state-of-the-art drift detection algorithms could achieve similar results by aligning their thresholds, providing a novel insight to explain how drift detection algorithms contribute to data stream learning. We noticed that a higher detection sensitivity improves accuracy for data streams with frequent distribution change. The evaluation results are showing that drift thresholds should not be fixed during stream learning. Rather, they should adjust dynamically based on the prevailing conditions of the data stream.

Using an ensemble of neural networks with consistency regularization is effective for improving performance and stability of deep learning, compared to the case of a single network. In this paper, we present a semi-supervised Deep Coupled Ensemble (DCE) model, which contributes to ensemble learning and classification landmark exploration for better locating the final decision boundaries in the learnt latent space. First, multiple complementary consistency regularizations are integrated into our DCE model to enable the ensemble members to learn from each other and themselves, such that training experience from different sources can be shared and utilized during training. Second, in view of the possibility of producing incorrect predictions on a number of difficult instances, we adopt class-wise mean feature matching to explore important unlabeled instances as classification landmarks, on which the model predictions are more reliable. Minimizing the weighted conditional entropy on unlabeled data is able to force the final decision boundaries to move away from important training data points, which facilitates semi-supervised learning. Ensemble members could eventually have similar performance due to consistency regularization, and thus only one of these members is needed during the test stage, such that the efficiency of our model is the same as the non-ensemble case. Extensive experimental results demonstrate the superiority of our proposed DCE model over existing state-of-the-art semi-supervised learning methods.

Network protocol classification plays an important role in modern network security and fine-grained management architectures. The state-of-the-art network protocol classification methods aim to take the advantages of flow statistical features and machine learning techniques. However the classification performance is severely affected by limited supervised information and unknown network protocols. In this paper, a novel semi-supervised learning method is proposed to solve the problem of unknown protocols in the crucial situation of a small labeled training sample set. The proposed method possesses the superior capability of detecting unknown samples generated by unknown protocols with the help of flow correlation information and semi-supervised clustering ensemble learning to boost the classification performance. A theoretical analysis is provided to confirm the effectiveness of the proposed method. Moreover, the comprehensive performance evaluation conducted on real-world network protocols datasets shows that the proposed method is significantly better than the existing methods in the critical network environment.