False Labels Research Articles

A three-stage learning approach to cross-domain person re-identification

Cross-domain person re-identification (Re-id) is far less discussed than supervised Re-id, since there are not labeled data that can be used to guide the person Re-id. Even though some works have been done on unsupervised Re-id, cross-domain Re-id is much challenging, and needs further research works. In this paper, without background transfer and other data enhancement of source domain training set, we propose a three-stage cross-domain Re-id model which considers the domain adaptation, self-supervised clustering re-training and joint loss training in one schema. In order to improve the adaptability of the model trained on the source data, cross-domain and cross-camera losses are considered. Then we do the self-supervised learning of clustering re-training in target dataset to learn the discriminative features in the target domain on the basis of the pre-trained model. Finally, we train the target data using the pseudo labels obtained by clustering with joint metric learning and representation learning. In particular, we replace the classic cross-entropy loss with label smoothing regularization loss so as to reduce fitting of false pseudo labels. Performance evaluation on benchmark datasets demonstrates the effectiveness of the proposed approach.

Loss Decomposition and Centroid Estimation for Positive and Unlabeled Learning.

This paper studies Positive and Unlabeled learning (PU learning), of which the target is to build a binary classifier where only positive data and unlabeled data are available for classifier training. To deal with the absence of negative training data, we first regard all unlabeled data as negative examples with false negative labels, and then convert PU learning into the risk minimization problem in the presence of such one-side label noise. Specifically, we propose a novel PU learning algorithm dubbed "Loss Decomposition and Centroid Estimation" (LDCE). By decomposing the loss function of corrupted negative examples into two parts, we show that only the second part is affected by the noisy labels. Thereby, we may estimate the centroid of corrupted negative set via an unbiased way to reduce the adverse impact of such label noise. Furthermore, we propose the "Kernelized LDCE" (KLDCE) by introducing the kernel trick, and show that KLDCE can be easily solved by combining Alternative Convex Search (ACS) and Sequential Minimal Optimization (SMO). Theoretically, we derive the generalization error bound which suggests that the generalization risk of our model converges to the empirical risk with the order of O(1/√k+1/√{n-k}+1/√n) ( n and k are the amounts of training data and positive data correspondingly). Experimentally, we conduct intensive experiments on synthetic dataset, UCI benchmark datasets and real-world datasets, and the results demonstrate that our approaches (LDCE and KLDCE) achieve the top-level performance when compared with both classic and state-of-the-art PU learning methods.

Robust Electricity Theft Detection Against Data Poisoning Attacks in Smart Grids

Data-driven electricity theft detectors rely on customers’ reported energy consumption readings to detect malicious behavior. One common implicit assumption in such detectors is the correct labeling of the training data. Unfortunately, these detectors are vulnerable against data poisoning attacks that assume false labels during training. This article addresses three major problems: What is the impact of data poisoning attacks on the detector’s performance? Which detector is more robust against data poisoning attacks, i.e., generalized or customer-specific detectors? How to improve the detector’s robustness against data poisoning attacks? Our investigations reveal that: (a) Shallow and deep learning-based detectors suffer from data poisoning attacks that may lead to a significant deterioration of detection rate of up to 17%. Furthermore, deep detectors offer 12% performance improvement over shallow detectors. (b) Generalized detectors present 4% performance improvement over customer-specific detectors even in the presence of data poisoning attacks. To enhance the detectors’ robustness against data poisoning attacks, we propose a sequential ensemble detector based on a deep auto-encoder with attention (AEA), gated recurrent units (GRUs), and feed forward neural networks. The proposed robust detector retains a stable detection performance that is deteriorated only by 1 – 3% in the presence of strong data poisoning attacks.

Optimizing Sector Ring Histogram of Oriented Gradients for human injured detection from drone images

Developing a system for emergency response and rescue to find injured people within images has been an interest to many researchers. The key challenge is to define a proper feature to describe the human body's appearance. Various features are often extracted from low-level data, such as texture and colour (Zhang et al., Sensors. 19(5):1005, 2020). One of the strong features is the Sector Ring Histogram of Oriented Gradients (SRHOG) that has been successfully applied to human detection tasks. Despite good accuracy in finding humans, an SRHOG detection method produces a large amount of false-negative labels. Locating an injured body after a disaster in drone images using an imaging camera remained a challenge. This study presents a new extension to SRHOG, so-called AdSRHOG, to reduce the number of false labels. In our approach, the gradient filters used by SRHOG defined adaptively depending on the corresponding pixel location The proposed feature was used the Support Vector Machine (SVM) algorithm to detect humans on drone images. The experiments showed a significant improvement of up to 54.3% in reducing the false labels. It was also found that the overall accuracy of the human detection process had a notable improvement of 13.1% over a traditional SRHOG detection technique.

Partial multi-label learning with mutual teaching

Partial Multi-label Learning (PML) tackles the problem where each training instance is associated with a set of candidate labels that include both the relevant ground-truth labels and irrelevant false positive labels. Most of the existing PML methods try to iteratively update the confidence of each candidate label, while the estimated label confidence may be not reliable due to the cumulative error induced in the confidence updating process, especially when false positive labels dominate. In this paper, we propose a simple yet effective model called PML-MT (Partial Multi-label Learning with Mutual Teaching), in which a couple of prediction networks as well as the corresponding teacher networks are adopted to learn collaboratively and teach each other throughout the training process. Specially, the proposed PML-MT model iteratively refines the label confidence matrix through a couple of self-ensemble teacher networks and trains two prediction networks simultaneously in a mutual teaching manner. Moreover, we propose a novel regularization term to further exploit label correlations from the outputs of the prediction networks under the supervision of the refined label confidence matrix. In addition, a co-regularization term is introduced to maximize the agreement on the outputs of the couple prediction networks, so that the predictions of each network would be more reliable. Extensive experiments on synthesized and real-world PML datasets demonstrate that the proposed approach outperforms the state-of-the-art counterparts.

Correcting and reweighting false label masks in brain tumor segmentation.

Recently, brain tumor segmentation has made important progress. However, the quality of manual labels plays an important role in the performance, while in practice, it could vary greatly and in turn could substantially mislead the learning process and decrease the accuracy. We need to design a mechanism to combine label correction and sample reweighting to improve the effectiveness of brain tumor segmentation. We propose a novel sample reweighting and label refinement method, and a novel three-dimensional (3D) generative adversarial network (GAN) is introduced to combine these two models into an united framework. Extensive experiments on the BraTS19 dataset have demonstrated that our approach obtains competitive results when compared with other state-of-the-art approaches when handling the false labels in brain tumor segmentation. The 3D GAN-based approach is an effective approach to handle false label masks by simultaneously applying label correction and sample reweighting. Our method is robust to variations in tumor shape and background clutter.

Weakly-Supervised Cross-Domain Adaptation for Endoscopic Lesions Segmentation

Weakly-supervised learning has attracted growing research attention on medical lesions segmentation due to significant saving in pixel-level annotation cost. However, 1) most existing methods require effective prior and constraints to explore the intrinsic lesions characterization, which only generates incorrect and rough prediction; 2) they neglect the underlying semantic dependencies among weakly-labeled target enteroscopy diseases and fully-annotated source gastroscope lesions, while forcefully utilizing untransferable dependencies leads to the negative performance. To tackle above issues, we propose a new weakly-supervised lesions transfer framework, which can not only explore transferable domain-invariant knowledge across different datasets, but also prevent the negative transfer of untransferable representations. Specifically, a Wasserstein quantified transferability framework is developed to highlight wide-range transferable contextual dependencies, while neglecting the irrelevant semantic characterizations. Moreover, a novel self-supervised pseudo label generator is designed to equally provide confident pseudo pixel labels for both hard-to-transfer and easy-to-transfer target samples. It inhibits the enormous deviation of false pseudo pixel labels under the self-supervision manner. Afterwards, dynamically-searched feature centroids are aligned to narrow category-wise distribution shift. Comprehensive theoretical analysis and experiments show the superiority of our model on the endoscopic dataset and several public datasets.

EVALUATING SECTOR RING HISTOGRAM OF ORIENTED GRADIENTS FILTER IN LOCATING HUMANS WITHIN UAV IMAGES

Abstract. Developing systems to find injured people quickly after natural disasters is an important topic. In recent years, special attention has been paid to the use of UAV images for this purpose. In this regard, an accurate and strong feature is required. It is shown that the Sector Ring Histogram of Oriented Gradients, is a feature very much independent from rotation and scale. The aim of this paper is to evaluate the performance of a human detection algorithm which is based on this strong feature. Experiments carried out suggest that using SRHOG feature humans can be detected with an accuracy of 73.69%. However, despite giving good accuracy, SRHOG results contain more than 33.33 % false labels.

Speech cues to deception in bilinguals.

Acoustic cues to deception on a picture naming task were analyzed in three groups of English speakers: monolinguals, bilinguals with English as their first language (English-L1), and bilinguals with English as a second language (English-L2). Results revealed that all participants had longer reaction times when generating falsehoods than when producing truths, and that the effect was more robust for English-L2 bilinguals than for the other two groups. Articulation rate was higher for all groups when producing lies. Mean fundamental frequency and intensity cues were not reliable cues to deception, but there was lower variance in both of these parameters when generating false vs. true labels for all participants. Results suggest that naming latency was the only cue to deception that differed by language background. These findings broadly support the cognitive-load theory of deception, suggesting that a combination of producing deceptive speech and using a second language puts an extra load on the speaker.

Partial Multi-Label Learning via Credible Label Elicitation.

Partial multi-label learning (PML) deals with the problem where each training example is associated with an overcomplete set of candidate labels, among which only some candidate labels are valid. The task of PML naturally arises in learning scenarios with inaccurate supervision, and the goal is to induce a multi-label predictor which can assign a set of proper labels for unseen instance. The PML training procedure is prone to be misled by false positive labels concealed in the candidate label set, which serves as the major modeling difficulty for partial multi-label learning. In this paper, a novel two-stage PML approach is proposed which works by eliciting credible labels from the candidate label set for model induction. In the first stage, the labeling confidence of candidate label for each PML training example is estimated via iterative label propagation. In the second stage, by utilizing credible labels with high labeling confidence, multi-label predictor is induced via pairwise label ranking coupled with virtual label splitting or maximum a posteriori (MAP) reasoning. Experimental studies show that the proposed approach can achieve highly competitive generalization performance by excluding most false positive labels from the training procedure via credible label elicitation.

Personal Stress-Level Clustering and Decision-Level Smoothing to Enhance the Performance of Ambulatory Stress Detection With Smartwatches

Researchers strive hard to develop effective ways to detect and cope with enduring high-level daily stress as early as possible to prevent serious health consequences. Although research has traditionally been conducted in laboratory settings, a set of new studies have recently begun to be conducted in ecological environments with unobtrusive wearable devices. Since patterns of stress are ideographic, person-independent models have generally lower accuracies. On the contrary, person-specific models have higher accuracies but they require a long-term data collection period. In this study, we developed a hybrid approach of personal level stress clustering by using baseline stress self-reports to increase the success of person-independent models without requiring a substantial amount of personal data. We further added decision level smoothing to our unobtrusive smartwatch based stress level differentiation system to increase the performance by correcting false labels assigned by the machine learning algorithm. In order to test and evaluate our system, we collected physiological data from 32 participants of a summer school with wrist-worn unobtrusive wearable devices. This event is comprised of baseline, lecture, exam and recovery sessions. In the recovery session, a stress management method was applied to alleviate the stress of the participants. The perceived stress in the form of NASA-TLX questionnaires collected from the users as self-reports and physiological stress levels extracted using wearable sensors are examined separately. By using our system, we were able to differentiate the 3-levels of stress successfully. We further substantially increase our performance by personal stress level clustering and by applying high-level accuracy calculation and decision level smoothing methods. We also demonstrated the success of the stress reduction methods by analyzing physiological signals and self-reports.

Disambiguation-free partial label learning

Partial label learning is an important weakly supervised machine learning framework. In partial label learning, each object is described by a single instance in the input space; however, in the output space, it is associated with a set of candidate labels among which only one is valid. An intuitive strategy is to disambiguate candidate labels, but this strategy tends to be misled by false positive labels; therefore, new disambiguation-free approaches need to be considered. In this paper, several algorithms are reviewed from the perspective of disambiguation and disambiguation-free strategies. First, the problem definition on partial label learning and its relationship with other related learning frameworks are given. Second, several representative partial label learning algorithms via the disambiguation strategy are introduced. Third, two of our proposed disambiguation-free algorithms are presented. Finally, the summary of this paper is given and possible future investigations on partial label learning are briefly discussed.

A Deep Learning Method for Change Detection in Synthetic Aperture Radar Images

With the rapid development of various technologies of satellite sensor, synthetic aperture radar (SAR) image has been an import source of data in the application of change detection. In this paper, a novel method based on a convolutional neural network (CNN) for SAR image change detection is proposed. The main idea of our method is to generate the classification results directly from the original two SAR images through a CNN without any preprocessing operations, which also eliminate the process of generating the difference image (DI), thus reducing the influence of the DI on the final classification result. In CNN, the spatial characteristics of the raw image can be extracted and captured by automatic learning and the results with stronger robustness can be obtained. The basic idea of the proposed method includes three steps: it first produces false labels through unsupervised spatial fuzzy clustering. Then we train the CNN through proper samples that are selected from the samples with false labels. Finally, the final detection results are obtained by the trained convolutional network. Although training the convolutional network is a supervised learning fashion, the whole process of the algorithm is an unsupervised process without priori knowledge. The theoretical analysis and experimental results demonstrate the validity, robustness, and potential of our algorithm in simulated and real data sets. In addition, we try to apply our algorithm to the change detection of heterogeneous images, which also achieves satisfactory results.

Partial Multi-Label Learning via Credible Label Elicitation

In partial multi-label learning (PML), each training example is associated with multiple candidate labels which are only partially valid. The task of PML naturally arises in learning scenarios with inaccurate supervision, and the goal is to induce a multi-label predictor which can assign a set of proper labels for unseen instance. To learn from PML training examples, the training procedure is prone to be misled by the false positive labels concealed in candidate label set. In light of this major difficulty, a novel two-stage PML approach is proposed which works by eliciting credible labels from the candidate label set for model induction. In this way, most false positive labels are expected to be excluded from the training procedure. Specifically, in the first stage, the labeling confidence of candidate label for each PML training example is estimated via iterative label propagation. In the second stage, by utilizing credible labels with high labeling confidence, multi-label predictor is induced via pairwise label ranking with virtual label splitting or maximum a posteriori (MAP) reasoning. Extensive experiments on synthetic as well as real-world data sets clearly validate the effectiveness of credible label elicitation in learning from PML examples.

Limited Gradient Descent: Learning With Noisy Labels

Label noise may affect the generalization of classifiers, and the effective learning of main patterns from samples with noisy labels is an important challenge. Recent studies have shown that deep neural networks tend to prioritize the learning of simple patterns over the memorization of noise patterns. This suggests a possible method to search for the best generalization that learns the main pattern until the noise begins to be memorized. Traditional approaches often employ a clean validation set to find the best stop timing of learning, i.e., early stopping. However, the generalization performance of such methods relies on the quality of validation sets. Further, in practice, a clean validation set is sometimes difficult to obtain. To solve this problem, we propose a method that can estimate the optimal stopping timing without a clean validation set, called limited gradient descent. We modified the labels of a few samples in a noisy dataset to obtain false labels and to create a reverse pattern. By monitoring the learning progress of the noisy and reverse samples, we can determine the stop timing of learning. In this paper, we also theoretically provide some necessary conditions on learning with noisy labels. Experimental results on CIFAR-10 and CIFAR-100 datasets demonstrate that our approach has a comparable generalization performance to methods relying on a clean validation set. Thus, on the noisy Clothing-1M dataset, our approach surpasses methods that rely on a clean validation set.

3D Deep Learning for Automatic Brain MR Tumor Segmentation with T-Spline Intensity Inhomogeneity Correction

Automatic segmentation of brain tumor data is a herculean task for medical applications, particularly in cancer diagnosis. This paper emulates some challenging issues such as noise sensitivity, partial volume averaging, intensity inhomogeneity, inter-slice intensity variations, and intensity non-standardization. This paper intends a novel N3T-spline intensity inhomogeneity correction for bias field correction and the three dimension convolutional neural network (3DCNN) for automatic segmentation. The proposed work consists of four stages (i) pre-processing, (ii) feature extraction (iii) automatic segmentation and (iv) post-processing. In the pre-processing step, novel nonparametric non-uniformity normalization (N3) based T-spline approach is proposed to correct the bias field distortion, which recedes the noises and intensity variations. The extended gray level co-occurrence matrix (EGLCM) is a feature extraction technique, from which the texture patches more suitable for brain tumor segmentation can be extracted. The proposed 3DCNN automatically segments the brain tumor and divides the discrete abnormal tissues from the raw data and EGLCM features. Finally, a simple threshold scheme is adapted on the segmented result to correct the false labels and eliminate the 3D connected small regions. The simulation results in the proposed segmentation procedure could acquire competitive performance as compared with the existing procedure for the BRATS 2015 dataset.

MEMPERDEBAT WUJÛDIYAH SYEIKH HAMZAH FANSURI (Kajian Hermeneutik atas Karya Sastra Hamzah Fansuri)

This paper is to analyze the causes of different opinions emerge regarding the provision of false labels, zindîq, pantheism, and similar for the teachings of Hamzah Fansuri’s wujûdiyah, so that it should be prior studies in accordance with the nature wujûdiyah itself with Sufi’s taught that institutionalized in the literary works with hermeneutic approach. Further, this research studies on the basis of what scholars have criticized this Sufi‘s teachings wujûdiyah and vice versa, on the basis of what scholars have given their support to this teaching

Disambiguation-Free Partial Label Learning

In partial label learning, each training example is associated with a set of candidate labels among which only one is the ground-truth label. The common strategy to induce predictive model is trying to disambiguate the candidate label set, i.e., differentiating the modeling outputs of individual candidate labels. Specifically, disambiguation by differentiation can be conducted either by identifying the ground-truth label iteratively or by treating each candidate label equally. Nonetheless, the disambiguation strategy is prone to be misled by the false positive labels co-occurring with ground-truth label. In this paper, a new partial label learning strategy is studied which refrains from conducting disambiguation. Specifically, by adapting error-correcting output codes (ECOC), a simple yet effective approach named Pl-ecoc is proposed by utilizing candidate label set as an entirety . During training phase, to build binary classifier w.r.t. each column coding, any partially labeled example will be regarded as a positive or negative training example only if its candidate label set entirely falls into the coding dichotomy. During testing phase, class label for the unseen instance is determined via loss-based decoding which considers binary classifiers’ empirical performance and predictive margin. Extensive experiments show that Pl-ecoc performs favorably against state-of-the-art partial label learning approaches.

Risk Factors for Non-communicable Diseases in Vietnam: A Focus on Pesticides

Agent Orange, which was sprayed in southern Vietnam by the American government, was the main source of dioxin exposure in Vietnam. From the early 1990s, agriculture of Vietnam has attained advances under intensive cultivation. Both production and yields per crop has raised significantly at the farm level, but, on the other hand, the quantity of pesticides used in agriculture has been increased in the absence of regulations and good practices. Illegal business of pesticides with false labels, as well as marketing of expired or poor quality products in stores without license are so popular in Vietnam. Misuse and improper use in agriculture in Vietnam has led to a variety of problems, such as environment pollution (including food producing animals) and adverse health impact on animals and humans. Open dumpsites worsen the general scenario. Similar to the environmental exposure, human exposure to DDT in Vietnam was ranked among the highest worldwide, with recognized effects. Exposed communities have to face birth defects, health disorders and non-communicable diseases (NCDs), from metabolic syndrome, asthma, infertility and other reproductive disorders through to diabetes, obesity, cardiovascular and neurodegenerative diseases, and cancer. A common feature of many chronic disorders and NCDs is metabolic disruption: environmental chemical factors disturb cellular homeostasis, thus affecting the ability of the body to restore a functional internal environment. Among these, endocrine disrupting pesticides can interfere with the action of hormones including metabolic hormones, and are likely to represent the main concern for developmentally-induced NCDs. Since pesticides are often persistent and bio-accumulate in the food chain through the living environment of food-producing organisms, this paper discusses relevant aspects of risk assessment, risk communication and risk management.

Can chemometrics protect pumpkin seed oil buyers from false Styrian PGI labels?

A group of scientists from Loeben Universtity is proposing the use of chemometrics in the detection of authenticity of Styrian PGI pumpkin seed oil. They have shown that pumpkin seeds from Austria and foreign countries differ in the composition of various element traces. Based on that they are able to differentiate between false and authentic PGI pumpkin seed using mathematical modeling.