Search

Fluxapyroxad (Figure 1), a broad-spectrum pyrazole-carboxamide fungicide, is commonly employed for the treatment of a large variety of commercial crops. More specifically, the application of Fluxap...

The success of training accurate models strongly depends on the availability of a sufficient collection of precisely labeled data. However, real-world datasets contain erroneously labeled data samples that substantially hinder the performance of machine learning models. Meanwhile, well-labeled data is usually expensive to obtain and only a limited amount is available for training. In this paper, we consider the problem of training a robust model by using large-scale noisy data in conjunction with a small set of clean data. To leverage the information contained via the clean labels, we propose a novel self-paced robust learning algorithm (SPRL) that trains the model in a process from more reliable (clean) data instances to less reliable (noisy) ones under the supervision of well-labeled data. The self-paced learning process hedges the risk of selecting corrupted data into the training set. Moreover, theoretical analyses on the convergence of the proposed algorithm are provided under mild assumptions. Extensive experiments on synthetic and real-world datasets demonstrate that our proposed approach can achieve a considerable improvement in effectiveness and robustness to existing methods.

With the advance of sensor technologies, the Multivariate Time Series classification (MTSC) problem, perhaps one of the most essential problems in the time series data mining domain, has continuously received a significant amount of attention in recent decades. Traditional time series classification approaches based on Bag-of-Patterns or Time Series Shapelet have difficulty dealing with the huge amounts of feature candidates generated in high-dimensional multivariate data but have promising performance even when the training set is small. In contrast, deep learning based methods can learn low-dimensional features efficiently but suffer from a shortage of labelled data. In this paper, we propose a novel MTSC model with an attentional prototype network to take the strengths of both traditional and deep learning based approaches. Specifically, we design a random group permutation method combined with multi-layer convolutional networks to learn the low-dimensional features from multivariate time series data. To handle the issue of limited training labels, we propose a novel attentional prototype network to train the feature representation based on their distance to class prototypes with inadequate data labels. In addition, we extend our model into its semi-supervised setting by utilizing the unlabeled data. Extensive experiments on 18 datasets in a public UEA Multivariate time series archive with eight state-of-the-art baseline methods exhibit the effectiveness of the proposed model.

In many high-stakes applications of machine learning models, outputting only predictions or providing statistical confidence is usually insufficient to gain trust from end users, who often prefer a transparent reasoning paradigm. Despite the recent encouraging developments on deep networks for sequential data modeling, due to the highly recursive functions, the underlying rationales of their predictions are difficult to explain. Thus, in this paper, we aim to develop a sequence modeling approach that explains its own predictions by breaking input sequences down into evidencing segments (i.e., sub-sequences) in its reasoning. To this end, we build our model upon convolutional neural networks, which, in their vanilla forms, associates local receptive fields with outputs in an obscure manner. To unveil it, we resort to case-based reasoning, and design prototype modules whose units (i.e., prototypes) resemble exemplar segments in the problem domain. Each prediction is obtained by combining the comparisons between the prototypes and the segments of an input. To enhance interpretability, we propose a training objective that delicately adapts the distribution of prototypes to the data distribution in latent spaces, and design an algorithm to map prototypes to human-understandable segments. Through extensive experiments in a variety of domains, we demonstrate that our model can achieve high interpretability generally, together with a competitive accuracy to the state-of-the-art approaches.

To accurate modulate the phase of the incoming light, the backplane aberration of a spatial light modulator (SLM) needs to be measured and compensated. In this paper, we develop an interferometric method to calibrate the static aberration. In our method, a Michelson interferometer was constructed and the SLM itself was used to produce the random phase shift that we need. In addition, the phase demodulation method based on matrix VU factorization (VU) and phase unwrapping algorithm based on derivative Zernike polynomial fitting (DZPT) are adopted to get the phase profile of the static aberration. Experimental result shows that our proposed method can get a pretty good compensation result.

Despite the great progress made by deep neural networks in the semantic segmentation task, traditional neural-network-based methods typically suffer from a shortage of large amounts of pixel-level annotations. Recent progress in few-shot semantic segmentation tackles the issue by only a few pixel-level annotated examples. However, these few-shot approaches cannot easily be applied to multi-way or weak an-notation settings. In this paper, we advance the few-shot segmentation paradigm towards a scenario where image-level an-notations are available to help the training process of a few pixel-level annotations. Our key idea is to learn a better prototype representation of the class by fusing the knowledge from the image-level labeled data. Specifically, we propose a new framework, called PAIA, to learn the class prototype representation in a metric space by integrating image-level annotations. Furthermore, by considering the uncertainty of pseudo-masks, a distilled soft masked average pooling strategy is designed to handle distractions in image-level annotations. Extensive empirical results on two datasets show superior performance of PAIA.

Measuring document similarity plays an important role in natural language processing tasks. Most existing document similarity approaches suffer from the information gap caused by context and vocabulary mismatches when comparing varying-length texts. In this paper, we propose an unsupervised concept representation learning approach to address the above issues. Specifically, we propose a novel Concept Generation Network (CGNet) to learn concept representations from the perspective of the entire text corpus. Moreover, a concept-based document matching method is proposed to leverage advances in the recognition of local phrase features and corpus-level concept features. Extensive experiments on real-world data sets demonstrate that new method can achieve a considerable improvement in comparing length-varying texts. In particular, our model achieved 6.5% better F1 Score compared to the best of the baseline models for a concept-project benchmark dataset.

Background: This study aimed to develop and validate nomograms for predicting overall survival (OS) and cancer-specific survival (CSS) in small intestinal gastrointestinal stromal tumors (SI GISTs). Methods: Patients diagnosed with SI GISTs were retrieved from the Surveillance, Epidemiology, and End Results (SEER) database and further randomly divided into training and validating cohorts. Univariate and multivariate Cox analyses were conducted in the training set to determine independent prognostic factors to build nomograms for predicting 3- and 5-year OS and CSS. The performance of the nomograms was assessed by using the concordance index (C-index), calibration plot, and the area under the receiver operating characteristic curve (AUC). Results: Data of a total of 776 patients with SI GISTs were retrospectively collected from the SEER database. The OS nomogram was constructed based on age, surgery, imatinib treatment, and American Joint Committee for Cancer (AJCC) stage, while the CSS nomogram incorporated age, surgery, tumor grade, and AJCC stage. In the training set, the C-index for the OS nomogram was 0.773 (95% confidence interval [95% CI]: 0.722–0.824) and for the CSS nomogram 0.806 (95% CI: 0.757–0.855). In the internal validation cohort, the C-index for the OS nomogram was 0.741, while for the CSS nomogram, it was 0.819. Well-corresponded calibration plots both in OS and CSS nomogram models were noticed. The comparisons of AUC values showed that the established nomograms exhibited superior discrimination power than the 7th Tumor-Node-Metastasis staging system. Conclusion: Our nomogram can effectively predict 3- and 5-year OS and CSS in patients with SI GISTs, and its use can help improve the accuracy of personalized survival prediction and facilitate to provide constructive therapeutic suggestions.

Question routing (QR) aims at recommending newly posted questions to the potential answerers who are most likely to answer the questions. The existing approaches that learn users' expertise from their past question-answering activities usually suffer from challenges in two aspects: 1) multi-faceted expertise and 2) temporal dynamics in the answering behavior. This paper proposes a novel temporal context-aware model in multiple granularities of temporal dynamics that concurrently address the above challenges. Specifically, the temporal context-aware attention characterizes the answerer's multi-faceted expertise in terms of the questions' semantic and temporal information simultaneously. Moreover, the design of the multi-shift and multi-resolution module enables our model to handle temporal impact on different time granularities. Extensive experiments on six datasets from different domains demonstrate that the proposed model significantly outperforms competitive baseline models.

The uncertainty measurement of classified results is especially important in areas requiring limited human resources for higher accuracy. For instance, data-driven algorithms diagnosing diseases need accurate uncertainty score to decide whether additional but limited quantity of experts are needed for rectification. However, few uncertainty models focus on improving the performance of text classification where human resources are involved. To achieve this, we aim at generating accurate uncertainty score by improving the confidence of winning scores. Thus, a model called MSD, which includes three independent components as ``mix-up", ``self-ensembling", ``distinctiveness score", is proposed to improve the accuracy of uncertainty score by reducing the effect of overconfidence of winning score and considering the impact of different categories of uncertainty simultaneously. MSD can be applied with different Deep Neural Networks. Extensive experiments with ablation setting are conducted on four real-world datasets, on which, competitive results are obtained.

With the increase of temporal data availability, time series classification has drawn a lot of attention in the literature because of its wide spectrum of applications in diverse domains (e.g., healthcare, bioinformatics and finance), ranging from human activity recognition to financial pattern identification. While significant progress has been made to solve time series classification problem, the success of such methods relies on data sufficiency, and may not well capture the quality embeddings when training triple instances are scarce and highly imbalance across classes. To address these challenges, we propose a prototype embedding framework-Deep Prototypical Networks (DPN), which leverages a main embedding space to capture the discrepancies of difference time series classes for alleviating data scarcity. In addition, we further augment DPN framework with a relationship-dependent masking module to automatically fuse relevant information with a distance metric learning process, which addresses the data imbalance issue and performs robust time series classification. Experimental results show significant and consistent improvements compared to state-of-the-art techniques.

Online purchase forecasting is of great importance in e-commerce platforms, which is the basis of how to present personalized interesting product lists to individual customers. However, predicting online purchases is not trivial as it is influenced by many factors including: (i) the complex temporal pattern with hierarchical inter-correlations; (ii) arbitrary category dependencies. To address these factors, we develop a Graph Multi-Scale Pyramid Networks (GMP) framework to fully exploit users' latent behavioral patterns with both multi-scale temporal dynamics and arbitrary inter-dependencies among product categories. In GMP, we first design a multi-scale pyramid modulation network architecture which seamlessly preserves the underlying hierarchical temporal factors--governing users' purchase behaviors. Then, we employ convolution recurrent neural network to encode the categorical temporal pattern at each scale. After that, we develop a resolution-wise recalibration gating mechanism to automatically re-weight the importance of each scale-view representations. Finally, a context-graph neural network module is proposed to adaptively uncover complex dependencies among category-specific purchases. Extensive experiments on real-world e-commerce datasets demonstrate the superior performance of our method over state-of-the-art baselines across various settings.

Predicting individuals personality traits with their social media profile has proved to be feasible, but researchers recently have run into bottlenecks on further improving the prediction accuracy. One major limitation is that existing studies failed to consider context information in predicting social media users’ behaviors. In this paper, we adopted the DIAMONDS situation theory in psychology to capture the context information in Facebook posts. To solve this issue, we proposed a novel situation-based feature interaction learning model. In this study, we extracted situation features according to the DIAMONDS lexicon and computed the interaction values between these situation features and the commonly used n-gram features at the post level. Features at the post level were aggregated up to the user level using the averaging strategy. A group lasso penalty was employed to enforce strong heredity in the model, which addressed the overfitting challenge introduced by the interaction features. Empirical tests on a large-scale data set have demonstrated the effectiveness of the proposed method.

The presence of data noise and corruptions recently invokes increasing attention on Robust Least Squares Regression (RLSR), which addresses the fundamental problem that learns reliable regression coefficients when response variables can be arbitrarily corrupted. Until now, several important challenges still cannot be handled concurrently: 1) exact recovery guarantee of regression coefficients 2) difficulty in estimating the corruption ratio parameter; and 3) scalability to massive dataset. This paper proposes a novel Robust Least squares regression algorithm via Heuristic Hard thresholding (RLHH), that concurrently addresses all the above challenges. Specifically, the algorithm alternately optimizes the regression coefficients and estimates the optimal uncorrupted set via heuristic hard thresholding without corruption ratio parameter until it converges. We also prove that our algorithm benefits from strong guarantees analogous to those of state-of-the-art methods in terms of convergence rates and recovery guarantees. We provide empirical evidence to demonstrate that the effectiveness of our new method is superior to that of existing methods in the recovery of both regression coefficients and uncorrupted sets, with very competitive efficiency.

Connecting the dots between diverse entities such as people and organizations is a vital task for forming hypotheses and uncovering latent relationships among complex and large datasets. Most existing approaches are designed to address the relationship of entities in news reports, documents and abstracts, but such approaches are not suitable for Twitter data streams due to their unstructured languages, short-length messages, heterogeneous features and massive size. The sheer size of Twitter data requires more efficient algorithms to connect the dots within a short period of time. We present a system that automatically constructs stories by connecting entities in Twitter datasets. An entity similarity model is designed that combines both traditional entity-related features and social network attributes and a novel story generation algorithm applied on the similarity model is proposed to cope with the massive Twitter datasets. Extensive experimental evaluations were conducted to demonstrate the effectiveness of this new approach.

Objective: To identify whether saliva supernatant miR-21 can serve as a novel potential biomarker in patients with esophageal cancer (EC). Methods: 32 patients with EC and 16 healthy controls were recruited in this study. Total RNA was extracted from saliva supernatant samples for measurement of miR-21 levels using RT-qPCR and relationships between miR-21 levels and clinical characteristics of EC patients were analyzed. Results: miR-21 was significantly higher in the EC than control groups. The sensitivity and specificity were 84.4% and 62.5% respectively. Supernatant miR-21 levels showed no significant correlation with cancer stage, differentiation and nodal metastasis. Conclusions: Saliva supernatant miR-21 may be a novel biomarker for EC.

Ozone is a main component of secondary pollutants of vehicle exhausts, and ozone exposure to children in urban areas may be associated with the development of asthma. However, little is known about the mechanism(s) by which ozone affects human airway epithelium and subsequent airway inflammation.Human bronchial epithelial cells were exposed to ozone at 0.16 mg/m(3) for varying periods. The concentrations of IL-1 and IL-6 secreted by the cells were measured by enzyme-linked immunosorbent assay (ELISA) and the contents of intracellular malondialdehyde (MDA) were determined. Furthermore, the conditional medium from the ozone-exposed cells was examined for stimulating human peripheral T lymphocytes from asthmatic patients and healthy subjects, and the production of cytokines was characterized by ELISA and quantitative real-time polymerase chain reaction (RT-PCR).Ozone stimulated the IL-1 and IL-6 production by BEAS-2B cells and its stimulatory effects were time dependent. Furthermore, ozone exposure significantly increased the levels of MDA in BEAS-2B cells, as compared with that of the cells without ozone exposure, in a time-dependent manner. In addition, the conditional medium from the cells exposed to ozone, but not control condition medium, significantly increased the relative levels of IL-1 mRNA transcripts in human peripheral T lymphocytes from asthmatic patients, but not healthy subjects. However, the conditional medium did not induce significantly increased levels of IL-2 production by peripheral T cells.Our data indicated that exposure to low levels of ozone for a short period induced increases in the pro-inflammatory markers and oxidative stress in epithelial cells, which might contribute to airway inflammation particularly in asthmatic children.

Early surgery is vital in the treatment of pancreatic cancer, which is often fatal. However, there is currently no useful noninvasive biomarker to screen for pancreatic cancer. Studies have documented that many salivary molecules can be used to detect systemic diseases. We investigated whether salivary miRNAs are useful biomarkers for detecting resectable pancreatic cancer. Using an Agilent microarray, salivary miRNAs were profiled from saliva samples of 8 patients with resectable pancreatic cancer and 8 healthy controls. Candidate biomarkers identified in the profiles were subjected to validation using quantitative PCR and an independent sample set of 40 patients with pancreatic cancer, 20 with benign pancreatic tumors (BPT), and 40 healthy controls. The validated salivary miRNA biomarkers were evaluated within three discriminatory categories: pancreatic cancer versus healthy control, pancreatic cancer versus BPT, and pancreatic cancer versus noncancer (healthy control + BPT). miR-3679-5p showed significant downregulation in the pancreatic cancer group within the three categories (P = 0.008, 0.007, and 0.002, respectively), whereas miR-940 showed significant upregulation in pancreatic cancer (P = 0.006, 0.004, and 0.0001, respectively). Logistic regression models combining the two salivary miRNAs were able to distinguish resectable pancreatic cancer within the three categories, showing sensitivities of 72.5%, 62.5%, and 70.0% and specificities of 70.0%, 80.0%, and 70.0%, respectively. Salivary miR-3679-5p and miR-940 possess good discriminatory power to detect resectable pancreatic cancer, with reasonable specificity and sensitivity. This report provides a new method for the early detection of pancreatic cancer and other systemic diseases by assessing salivary miRNAs. Cancer Prev Res; 8(2); 165–73. ©2014 AACR.

Turning Web customers from "window shoppers" into buyers demands an interactive sales model that informs them, gives them individualized attention, and helps to close the sale at the customer's request. Ideally, sales agents should have in-person meetings with all prospective customers. However, this may not be desirable or feasible, The next best thing is for sales agents to send promotional e-mails to their prospective customers. In this paper, we describe the development of a direct marketing system that uses augmented reality (AR) technology. A set of specially designed markers is used to calibrate the camera and track the motion of the markers for the augmentation of three dimensional product models. There is no special hardware required for this system except a PC camera (e.g., WebCam or ViCAM),.

The uncertainty measurement of classifiers' predictions is especially important in applications such as medical diagnoses that need to ensure limited human resources can focus on the most uncertain predictions returned by machine learning models. However, few existing uncertainty models attempt to improve overall prediction accuracy where human resources are involved in the text classification task. In this paper, we propose a novel neural-network-based model that applies a new dropout-entropy method for uncertainty measurement. We also design a metric learning method on feature representations, which can boost the performance of dropout-based uncertainty methods with smaller prediction variance in accurate prediction trials. Extensive experiments on real-world data sets demonstrate that our method can achieve a considerable improvement in overall prediction accuracy compared to existing approaches. In particular, our model improved the accuracy from 0.78 to 0.92 when 30\% of the most uncertain predictions were handed over to human experts in "20NewsGroup" data., Comment: Accepted by NAACL19