Methods based on supervised learning using annotations in an end-to-end fashion have been the state-of-the-art for classification problems. However, they may be limited in their generalization capability, especially in the low data regime. In this study, we address this issue using supervised contrastive learning combined with available metadata to solve multiple pretext tasks that learn a good representation of data. We apply our approach on respiratory sound classification. This task is suited for this setting as demographic information such as sex and age are correlated with presence of lung diseases, and learning a system that implicitly encode this information may better detect anomalies. Supervised contrastive learning is a paradigm that learns similar representations to samples sharing the same class labels and dissimilar representations to samples with different class labels. The feature extractor learned using this paradigm extract useful features from the data, and we show that it outperforms cross-entropy in classifying respiratory anomalies in two different datasets. We also show that learning representations using only metadata, without class labels, obtains similar performance as using cross entropy with those labels only. In addition, when combining class labels with metadata using multiple supervised contrastive learning, an extension of supervised contrastive learning solving an additional task of grouping patients within the same sex and age group, more informative features are learned. This work suggests the potential of using multiple metadata sources in supervised contrastive settings, in particular in settings with class imbalance and few data. Our code is released at https://github.com/ilyassmoummad/scl_icbhi2017
Recent findings show that pre-trained wav2vec 2.0 models are reliable feature extractors for various speaker characteristics classification tasks. We show that latent representations extracted at different layers of a pre-trained wav2vec 2.0 system can be used as features for binary classification to distinguish between children with Cleft Lip and Palate (CLP) and a healthy control group. The results indicate that the distinction between CLP and healthy voices, especially with latent representations from the lower and middle encoder layers, reaches an accuracy of 100%. We test the classifier to find influencing factors for classification using unseen out-of-domain healthy and pathologic corpora with varying characteristics: age, spoken content, and acoustic conditions. Cross-pathology and cross-healthy tests reveal that the trained classifiers are unreliable if there is a mismatch between training and out-of-domain test data in, e.g., age, spoken content, or acoustic conditions.
Graph Echo State Networks (GESN) have already demonstrated their efficacy and efficiency in graph classification tasks. However, semi-supervised node classification brought out the problem of over-smoothing in end-to-end trained deep models, which causes a bias towards high homophily graphs. We evaluate for the first time GESN on node classification tasks with different degrees of homophily, analyzing also the impact of the reservoir radius. Our experiments show that reservoir models are able to achieve better or comparable accuracy with respect to fully trained deep models that implement ad hoc variations in the architectural bias, with a gain in terms of efficiency.
We theoretically study the impact of differential privacy on fairness in classification. We prove that, given a class of models, popular group fairness measures are pointwise Lipschitz-continuous with respect to the parameters of the model. This result is a consequence of a more general statement on accuracy conditioned on an arbitrary event (such as membership to a sensitive group), which may be of independent interest. We use this Lipschitz property to prove a non-asymptotic bound showing that, as the number of samples increases, the fairness level of private models gets closer to the one of their non-private counterparts. This bound also highlights the importance of the confidence margin of a model on the disparate impact of differential privacy.
The Hyperspectral image (HSI) contains several hundred bands of the same region called the Ground Truth (GT). The bands are taken in juxtaposed frequencies, but some of them are noisily measured or contain no information. For the classification, the selection of bands, affects significantly the results of classification, in fact, using a subset of relevant bands, these results can be better than those obtained using all bands, from which the need to reduce the dimensionality of the HSI. In this paper, a categorization of dimensionality reduction methods, according to the generation process, is presented. Furthermore, we reproduce an algorithm based on mutual information (MI) to reduce dimensionality by features selection and we introduce an algorithm using mutual information and homogeneity. The two schemas are a filter strategy. Finally, to validate this, we consider the case study AVIRIS HSI 92AV3C. Keywords: Hyperspectrale images; classification; features selection; mutual information; homogeneity
Cardiovascular activities are directly related to the response of a body in a stressed condition. Stress, based on its intensity, can be divided into two types i.e. Acute stress (short-term stress) and Chronic stress (long-term stress). Repeated acute stress and continuous chronic stress may play a vital role in inflammation in the circulatory system and thus leads to a heart attack or to a stroke. In this study, we have reviewed commonly used machine learning classification techniques applied to different stress-indicating parameters used in stress monitoring devices. These parameters include Photoplethysmograph (PPG), Electrocardiographs (ECG), Electromyograph (EMG), Galvanic Skin Response (GSR), Heart Rate Variation (HRV), skin temperature, respiratory rate, Electroencephalograph (EEG) and salivary cortisol, used in stress monitoring devices. This study also provides a discussion on choosing a classifier, which depends upon a number of factors other than accuracy, like the number of subjects involved in an experiment, type of signals processing and computational limitations.
Very high-resolution (VHR) remote sensing (RS) image classification is the fundamental task for RS image analysis and understanding. Recently, transformer-based models demonstrated outstanding potential for learning high-order contextual relationships from natural images with general resolution (224x224 pixels) and achieved remarkable results on general image classification tasks. However, the complexity of the naive transformer grows quadratically with the increase in image size, which prevents transformer-based models from VHR RS image (500x500 pixels) classification and other computationally expensive downstream tasks. To this end, we propose to decompose the expensive self-attention (SA) into real and imaginary parts via discrete Fourier transform (DFT) and therefore propose an efficient complex self-attention (CSA) mechanism. Benefiting from the conjugated symmetric property of DFT, CSA is capable to model the high-order contextual information with less than half computations of naive SA. To overcome the gradient explosion in Fourier complex field, we replace the Softmax function with the carefully designed Logmax function to normalize the attention map of CSA and stabilize the gradient propagation. By stacking various layers of CSA blocks, we propose the Fourier Complex Transformer (FCT) model to learn global contextual information from VHR aerial images following the hierarchical manners. Universal experiments conducted on commonly used RS classification data sets demonstrate the effectiveness and efficiency of FCT, especially on very high-resolution RS images.
Self-supervised models trained with a contrastive loss such as CLIP have shown to be very powerful in zero-shot classification settings. However, to be used as a zero-shot classifier these models require the user to provide new captions over a fixed set of labels at test time. In many settings, it is hard or impossible to know if a new query caption is compatible with the source captions used to train the model. We address these limitations by framing the zero-shot classification task as an outlier detection problem and develop a conformal prediction procedure to assess when a given test caption may be reliably used. On a real-world medical example, we show that our proposed conformal procedure improves the reliability of CLIP-style models in the zero-shot classification setting, and we provide an empirical analysis of the factors that may affect its performance.
The universal-set naive Bayes classifier (UNB)~\cite{Komiya:13}, defined using likelihood ratios (LRs), was proposed to address imbalanced classification problems. However, the LR estimator used in the UNB overestimates LRs for low-frequency data, degrading the classification performance. Our previous study~\cite{Kikuchi:19} proposed an effective LR estimator even for low-frequency data. This estimator uses regularization to suppress the overestimation, but we did not consider imbalanced data. In this paper, we integrated the estimator with the UNB. Our experiments with imbalanced data showed that our proposed classifier effectively adjusts the classification scores according to the class balance using regularization parameters and improves the classification performance.
Neural networks became the standard technique for image classification throughout the last years. They are extracting image features from a large number of images in a training phase. In a following test phase, the network is applied to the problem it was trained for and its performance is measured. In this paper, we focus on image classification. The amount of visual data that is interpreted by neural networks grows with the increasing usage of neural networks. Mostly, the visual data is transmitted from the application side to a central server where the interpretation is conducted. If the transmission is disturbed, losses occur in the transmitted images. These losses have to be reconstructed using postprocessing. In this paper, we incorporate the widely applied bilinear and bicubic interpolation and the high-quality reconstruction Frequency-Selective Reconstruction (FSR) for the reconstruction of corrupted images. However, we propose to use Frequency-Selective Mesh-to-Grid Resampling (FSMR) for the joint reconstruction and resizing of corrupted images. The performance in terms of classification accuracy of EfficientNetB0, DenseNet121, DenseNet201, ResNet50 and ResNet152 is examined. Results show that the reconstruction with FSMR leads to the highest classification accuracy for most networks. Average improvements of up to 6.7 percentage points are possible for DenseNet121.
The current trend towards more renewable and sustainable energy generation leads to an increased interest in new energy management systems and the concept of a smart grid. One important aspect of this is local energy trading, which is an extension of existing electricity markets by including prosumers, who are consumers also producing electricity. Prosumers having a surplus of energy may directly trade this surplus with other prosumers, which are currently in demand. In this paper, we present an overview of the literature in the area of local energy trading. In order to provide structure to the broad range of publications, we identify key characteristics, define the various settings, and cluster the considered literature along these characteristics. We identify three main research lines, each with a distinct setting and research question. We analyze and compare the settings, the used techniques, and the results and findings within each cluster and derive connections between the clusters. In addition, we identify important aspects, which up to now have to a large extent been neglected in the considered literature and highlight interesting research directions, and open problems for future work.
Text classification is a very classic NLP task, but it has two prominent shortcomings: On the one hand, text classification is deeply domain-dependent. That is, a classifier trained on the corpus of one domain may not perform so well in another domain. On the other hand, text classification models require a lot of annotated data for training. However, for some domains, there may not exist enough annotated data. Therefore, it is valuable to investigate how to efficiently utilize text data from different domains to improve the performance of models in various domains. Some multi-domain text classification models are trained by adversarial training to extract shared features among all domains and the specific features of each domain. We noted that the distinctness of the domain-specific features is different, so in this paper, we propose to use a curriculum learning strategy based on keyword weight ranking to improve the performance of multi-domain text classification models. The experimental results on the Amazon review and FDU-MTL datasets show that our curriculum learning strategy effectively improves the performance of multi-domain text classification models based on adversarial learning and outperforms state-of-the-art methods.
Recently, the hyperspectral sensors have improved our ability to monitor the earth surface with high spectral resolution. However, the high dimensionality of spectral data brings challenges for the image processing. Consequently, the dimensionality reduction is a necessary step in order to reduce the computational complexity and increase the classification accuracy. In this paper, we propose a new filter approach based on information gain for dimensionality reduction and classification of hyperspectral images. A special strategy based on hyperspectral bands selection is adopted to pick the most informative bands and discard the irrelevant and noisy ones. The algorithm evaluates the relevancy of the bands based on the information gain function with the support vector machine classifier. The proposed method is compared using two benchmark hyperspectral datasets (Indiana, Pavia) with three competing methods. The comparison results showed that the information gain filter approach outperforms the other methods on the tested datasets and could significantly reduce the computation cost while improving the classification accuracy. Keywords: Hyperspectral images; dimensionality reduction; information gain; classification accuracy. Keywords: Hyperspectral images; dimensionality reduction; information gain; classification accuracy.
Over the past decades, the hyperspectral remote sensing technology development has attracted growing interest among scientists in various domains. The rich and detailed spectral information provided by the hyperspectral sensors has improved the monitoring and detection capabilities of the earth surface substances. However, the high dimensionality of the hyperspectral images (HSI) is one of the main challenges for the analysis of the collected data. The existence of noisy, redundant and irrelevant bands increases the computational complexity, induce the Hughes phenomenon and decrease the target's classification accuracy. Hence, the dimensionality reduction is an essential step to face the dimensionality challenges. In this paper, we propose a novel filter approach based on the maximization of the spectral interaction measure and the support vector machines for dimensionality reduction and classification of the HSI. The proposed Max Relevance Max Synergy (MRMS) algorithm evaluates the relevance of every band through the combination of spectral synergy, redundancy and relevance measures. Our objective is to select the optimal subset of synergistic bands providing accurate classification of the supervised scene materials. Experimental results have been performed using three different hyperspectral datasets: "Indiana Pine", "Pavia University" and "Salinas" provided by the "NASA-AVIRIS" and the "ROSIS" spectrometers. Furthermore, a comparison with the state of the art band selection methods has been carried out in order to demonstrate the robustness and efficiency of the proposed approach. Keywords: Hyperspectral images, remote sensing, dimensionality reduction, classification, synergic, correlation, spectral interaction information, mutual inform
Great progress has been made in point cloud classification with learning-based methods. However, complex scene and sensor inaccuracy in real-world application make point cloud data suffer from corruptions, such as occlusion, noise and outliers. In this work, we propose Point-Voxel based Adaptive (PV-Ada) feature abstraction for robust point cloud classification under various corruptions. Specifically, the proposed framework iteratively voxelize the point cloud and extract point-voxel feature with shared local encoding and Transformer. Then, adaptive max-pooling is proposed to robustly aggregate the point cloud feature for classification. Experiments on ModelNet-C dataset demonstrate that PV-Ada outperforms the state-of-the-art methods. In particular, we rank the $2^{nd}$ place in ModelNet-C classification track of PointCloud-C Challenge 2022, with Overall Accuracy (OA) being 0.865. Code will be available at https://github.com/zhulf0804/PV-Ada.
Band selection is a great challenging task in the classification of hyperspectral remotely sensed images HSI. This is resulting from its high spectral resolution, the many class outputs and the limited number of training samples. For this purpose, this paper introduces a new filter approach for dimension reduction and classification of hyperspectral images using information theoretic (normalized mutual information) and support vector machines SVM. This method consists to select a minimal subset of the most informative and relevant bands from the input datasets for better classification efficiency. We applied our proposed algorithm on two well-known benchmark datasets gathered by the NASA's AVIRIS sensor over Indiana and Salinas valley in USA. The experimental results were assessed based on different evaluation metrics widely used in this area. The comparison with the state of the art methods proves that our method could produce good performance with reduced number of selected bands in a good timing. Keywords: Dimension reduction, Hyperspectral images, Band selection, Normalized mutual information, Classification, Support vector machines
Hyperspectral imagery (HSI) one-class classification is aimed at identifying a single target class from the HSI by using only knowing positive data, which can significantly reduce the requirements for annotation. However, when one-class classification meets HSI, it is difficult for classifiers to find a balance between the overfitting and underfitting of positive data due to the problems of distribution overlap and distribution imbalance. Although deep learning-based methods are currently the mainstream to overcome distribution overlap in HSI multiclassification, few studies focus on deep learning-based HSI one-class classification. In this article, a weakly supervised deep HSI one-class classifier, namely, HOneCls, is proposed, where a risk estimator,the one-class risk estimator, is particularly introduced to make the fully convolutional neural network (FCN) with the ability of one class classification in the case of distribution imbalance. Extensive experiments (20 tasks in total) were conducted to demonstrate the superiority of the proposed classifier.
Nowadays, the hyperspectral remote sensing imagery HSI becomes an important tool to observe the Earth's surface, detect the climatic changes and many other applications. The classification of HSI is one of the most challenging tasks due to the large amount of spectral information and the presence of redundant and irrelevant bands. Although great progresses have been made on classification techniques, few studies have been done to provide practical guidelines to determine the appropriate classifier for HSI. In this paper, we investigate the performance of four supervised learning algorithms, namely, Support Vector Machines SVM, Random Forest RF, K-Nearest Neighbors KNN and Linear Discriminant Analysis LDA with different kernels in terms of classification accuracies. The experiments have been performed on three real hyperspectral datasets taken from the NASA's Airborne Visible/Infrared Imaging Spectrometer Sensor AVIRIS and the Reflective Optics System Imaging Spectrometer ROSIS sensors. The mutual information had been used to reduce the dimensionality of the used datasets for better classification efficiency. The extensive experiments demonstrate that the SVM classifier with RBF kernel and RF produced statistically better results and seems to be respectively the more suitable as supervised classifiers for the hyperspectral remote sensing images. Keywords: hyperspectral images, mutual information, dimension reduction, Support Vector Machines, K-Nearest Neighbors, Random Forest, Linear Discriminant Analysis.
Automatic assessment of dysarthric speech is essential for sustained treatments and rehabilitation. However, obtaining atypical speech is challenging, often leading to data scarcity issues. To tackle the problem, we propose a novel automatic severity assessment method for dysarthric speech, using the self-supervised model in conjunction with multi-task learning. Wav2vec 2.0 XLS-R is jointly trained for two different tasks: severity classification and auxiliary automatic speech recognition (ASR). For the baseline experiments, we employ hand-crafted acoustic features and machine learning classifiers such as SVM, MLP, and XGBoost. Explored on the Korean dysarthric speech QoLT database, our model outperforms the traditional baseline methods, with a relative percentage increase of 1.25% for F1-score. In addition, the proposed model surpasses the model trained without ASR head, achieving 10.61% relative percentage improvements. Furthermore, we present how multi-task learning affects the severity classification performance by analyzing the latent representations and regularization effect.
With the increasing availability of large digitized fine art collections, automated analysis and classification of paintings is becoming an interesting area of research. However, due to domain specificity, implicit subjectivity, and pervasive nuances that vaguely separate art movements, analyzing art using machine learning techniques poses significant challenges. Residual networks, or variants thereof, are one the most popular tools for image classification tasks, which can extract relevant features for well-defined classes. In this case study, we focus on the classification of a selected painting 'Portrait of the Painter Charles Bruni' by Johann Kupetzky and the analysis of the performance of the proposed classifier. We show that the features extracted during residual network training can be useful for image retrieval within search systems in online art collections.