Search

Probabilistic relational PCA (PRPCA) can learn a projection matrix to perform dimensionality reduction for relational data. However, the results learned by PRPCA lack interpretability because each principal component is a linear combination of all the original variables. In this paper, we propose a novel model, called sparse probabilistic relational projection (SPRP), to learn a sparse projection matrix for relational dimensionality reduction. The sparsity in SPRP is achieved by imposing on the projection matrix a sparsity-inducing prior such as the Laplace prior or Jeffreys prior. We propose an expectation-maximization (EM) algorithm to learn the parameters of SPRP. Compared with PRPCA, the sparsity in SPRP not only makes the results more interpretable but also makes the projection operation much more efficient without compromising its accuracy. All these are verified by experiments conducted on several real applications.

Twitter sentiment analysis (TSA) has become a hot research topic in recent years. The goal of this task is to discover the attitude or opinion of the tweets, which is typically formulated as a machine learning based text classification problem. Some methods use manually labeled data to train fully supervised models, while others use some noisy labels, such as emoticons and hashtags, for model training. In general, we can only get a limited number of training data for the fully supervised models because it is very labor-intensive and time-consuming to manually label the tweets. As for the models with noisy labels, it is hard for them to achieve satisfactory performance due to the noise in the labels although it is easy to get a large amount of data for training. Hence, the best strategy is to utilize both manually labeled data and noisy labeled data for training. However, how to seamlessly integrate these two different kinds of data into the same learning framework is still a challenge. In this paper, we present a novel model, called emoticon smoothed language model (ESLAM), to handle this challenge. The basic idea is to train a language model based on the manually labeled data, and then use the noisy emoticon data for smoothing. Experiments on real data sets demonstrate that ESLAM can effectively integrate both kinds of data to outperform those methods using only one of them.

Hashing, which tries to learn similarity-preserving binary codes for data representation, has been widely used for efficient nearest neighbor search in massive databases due to its fast query speed and low storage cost. Because it is NP hard to directly compute the best binary codes for a given data set, mainstream hashing methods typically adopt a two-stage strategy. In the first stage, several projected dimensions of real values are generated. Then in the second stage, the real values will be quantized into binary codes by thresholding. Currently, most existing methods use one single bit to quantize each projected dimension. One problem with this single-bit quantization (SBQ) is that the threshold typically lies in the region of the highest point density and consequently a lot of neighboring points close to the threshold will be hashed to totally different bits, which is unexpected according to the principle of hashing. In this paper, we propose a novel quantization strategy, called double-bit quantization (DBQ), to solve the problem of SBQ. The basic idea of DBQ is to quantize each projected dimension into double bits with adaptively learned thresholds. Extensive experiments on two real data sets show that our DBQ strategy can significantly outperform traditional SBQ strategy for hashing.

Background Small pulmonary nodules are increasingly detected at an earlier stage and need to be removed via video-assisted thoracoscopic surgery (VATS). However, small pulmonary nodules are often difficult to locate during VATS and are typically nonvisible and nonpalpable on the lung surface. A variety of localization techniques have been developed. Here, we explored the application of an intraoperative body surface localization (IOBSL) and/or anatomical landmark localization (ALL) in minimally invasive surgery for small pulmonary nodules. Methods A total of 174 patients with small pulmonary nodules were divided into 3 groups: an IOBSL group, an ALL group, and an IOBSL+ALL group. VATS partial pneumonectomy was performed after the nodule localization, and the need for pulmonary segmentectomy/lobectomy and lymph node dissection was assessed according to the results of intraoperative rapid frozen section diagnosis. The duration, accuracy, and complications of each localization method were recorded and analyzed. Results ALL had shorter distance to the nodules (P=0.0282) but longer localization duration (P

The exponential increase in computation and memory complexity with the depth of network has become the main impediment to the successful application of graph neural networks (GNNs) on large-scale graphs like graphs with hundreds of millions of nodes. In this paper, we propose a novel neighbor sampling strategy, dubbed blocking-based neighbor sampling (BNS), for efficient training of GNNs on large-scale graphs. Specifically, BNS adopts a policy to stochastically block the ongoing expansion of neighboring nodes, which can reduce the rate of the exponential increase in computation and memory complexity of GNNs. Furthermore, a reweighted policy is applied to graph convolution, to adjust the contribution of blocked and non-blocked neighbors to central nodes. We theoretically prove that BNS provides an unbiased estimation for the original graph convolution operation. Extensive experiments on three benchmark datasets show that, on large-scale graphs, BNS is 2X~5X faster than state-of-the-art methods when achieving the same accuracy. Moreover, even on the small-scale graphs, BNS also demonstrates the advantage of low time cost.

AIM: To investigate the relationship between autophagy and apoptosis in photoinduced injuries in retinal pigment epithelium (RPE) cells and how Lycium barbarum polysaccharide (LBP) contributes to the increased of RPE cells to photoinduced autophagy. METHODS: In vitro cultures of human RPE strains (ARPE-19) were prepared and randomly divided into the blank control, model, low-dose LBP, middle-dose LBP, high-dose LBP, and 3-methyladenine (3MA) groups. The viability of the RPE cells and apoptosis levels in each group were tested through cell counting kit-8 (CCK8) method with a flow cytometer (Annexin V/PI double staining technique). The expression levels of LC3II, LC3I, and P62 proteins were detected with the immunofluorescence method. The expression levels of beclin1, LC3, P62, PI3K, P-mTOR, mTOR, P-Akt, and Akt proteins were tested through Western blot. RESULTS: LBP considerably strengthens cell viability and inhibits the apoptosis of RPE cells after photoinduction. The PI3K/Akt/mTOR signal pathway is activated because of the upregulation of the phosphorylation levels of Akt and mTOR proteins, and thus autophagy is inhibited. CONCLUSION: LBP can inhibit the excessive autophagy in RPE cells by activating the PI3K/Akt/mTOR signaling pathways and thereby protect RPE cells from photoinduced injuries.

The specific impacts of solid and micropapillary components on prognosis in lung adenocarcinoma remain unclear. Herein, we elucidated their distinct contributions to lung adenocarcinoma recurrence.Lung adenocarcinoma was classified into solid and micropapillary absent (S-M-); solid absent, micropapillary present (S-M+); micropapillary absent, solid present (S + M-); and solid and micropapillary present (S + M+). Cumulative incidence of recurrence (CIR) was calculated using competing risk analysis.Of 994 adenocarcinomas, 650 (65.4%) were classified as S-M-; 152 (15.3%), S-M+; 148 (14.9%), S + M-; and 44 (4.4%), S + M+. In total, 168 (16.9%) patients had recurrence; 16 (1.6%) died from other causes. S-M- had significantly lower CIR than other groups (S-M- vs. S-M+: P 0.001, S-M- vs. S + M-: P 0.001, S-M- vs. S + M+: P 0.001); S + M- had significantly higher CIR than S-M+ (P = 0.002). These differences remained significant in multivariable analysis. In stage IA, S-M- had significantly lower CIR than other groups (S-M- vs. S-M+: P = 0.006, S-M- vs. S + M-: P 0.001, S-M- vs. S + M+: P 0.001); S + M- and S + M+ had significantly higher CIR than S-M+ (P = 0.005, P = 0.008, respectively). These differences remained significant in multivariable analysis. CIR was not significantly different between S + M- and S-M+ subgroups.The presence of solid or micropapillary component (≥1%) was an independent risk factor for CIR; patients with solid component alone had a higher CIR than those with micropapillary component alone. In IA lung adenocarcinoma, patients with both solid and micropapillary components had a higher CIR than those with micropapillary component alone; the proportion of solid or micropapillary component was not associated with CIR.

Performance degradation caused by noise has been a long-standing challenge for speaker verification. Previous methods usually involve applying a denoising transformation to speaker embeddings or enhancing input features. Nevertheless, these methods are lossy and inefficient for speaker embedding. In this paper, we propose context- aware masking (CAM), a novel method to extract robust speaker embedding. CAM enables the speaker embedding network to "focus" on the speaker of interest and "blur" unrelated noise. The threshold of masking is dynamically controlled by an auxiliary context embedding that captures speaker and noise characteristics. Moreover, models adopting CAM can be trained in an end-to-end manner without using synthesized noisy-clean speech pairs. Our results show that CAM improves speaker verification performance in the wild by a large margin, compared to the baselines.

Due to its storage and retrieval efficiency, cross-modal hashing (CMH) has been widely used for cross-modal similarity search in many multimedia applications. According to the training strategy, existing CMH methods can be mainly divided into two categories: relaxation-based continuous methods and discrete methods. In general, the training of relaxation-based continuous methods is faster than that of discrete methods, but the accuracy of relaxation-based continuous methods is not satisfactory. On the contrary, the accuracy of discrete methods is typically better than that of the relaxation-based continuous methods, but the training of discrete methods is very time-consuming. In this paper, we propose a novel CMH method, called Discrete Latent Factor model-based cross-modal Hashing (DLFH), for cross modal similarity search. DLFH is a discrete method which can directly learn the binary hash codes for CMH. At the same time, the training of DLFH is efficient. Experiments show that the DLFH can achieve significantly better accuracy than existing methods, and the training time of DLFH is comparable to that of the relaxation-based continuous methods which are much faster than the existing discrete methods.

We aimed to develop a deep learning-based segmentation system for rapid on-site cytopathology evaluation (ROSE) to improve the diagnostic efficiency of endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) biopsy.A retrospective, multicenter, diagnostic study was conducted using 5345 cytopathological slide images from 194 patients who underwent EUS-FNA. These patients were from Nanjing Drum Tower Hospital (109 patients), Wuxi People's Hospital (30 patients), Wuxi Second People's Hospital (25 patients), and The Second Affiliated Hospital of Soochow University (30 patients). A deep convolutional neural network (DCNN) system was developed to segment cell clusters and identify cancer cell clusters with cytopathological slide images. Internal testing, external testing, subgroup analysis, and human-machine competition were used to evaluate the performance of the system.The DCNN system segmented stained cells from the background in cytopathological slides with an F1-score of 0·929 and 0·899-0·938 in internal and external testing, respectively. For cancer identification, the DCNN system identified images containing cancer clusters with AUCs of 0·958 and 0·948-0·976 in internal and external testing, respectively. The generalizable and robust performance of the DCNN system was validated in sensitivity analysis (AUC0·900) and was superior to that of trained endoscopists and comparable to cytopathologists on our testing datasets.The DCNN system is feasible and robust for identifying sample adequacy and pancreatic cancer cell clusters. Prospective studies are warranted to evaluate the clinical significance of the system.Jiangsu Natural Science Foundation; Nanjing Medical Science and Technology Development Funding; National Natural Science Foundation of China.

Non-convex models, like deep neural networks, have been widely used in machine learning applications. Training non-convex models is a difficult task owing to the saddle points of models. Recently, stochastic normalized gradient descent (SNGD), which updates the model parameter by a normalized gradient in each iteration, has attracted much attention. Existing results show that SNGD can achieve better performance on escaping saddle points than classical training methods like stochastic gradient descent (SGD). However, none of the existing studies has provided theoretical proof about the convergence of SNGD for non-convex problems. In this paper, we firstly prove the convergence of SNGD for non-convex problems. Particularly, we prove that SNGD can achieve the same computation complexity as SGD. In addition, based on our convergence proof of SNGD, we find that SNGD needs to adopt a small constant learning rate for convergence guarantee. This makes SNGD do not perform well on training large non-convex models in practice. Hence, we propose a new method, called stagewise SNGD (S-SNGD), to improve the performance of SNGD. Different from SNGD in which a small constant learning rate is necessary for convergence guarantee, S-SNGD can adopt a large initial learning rate and reduce the learning rate by stage. The convergence of S-SNGD can also be theoretically proved for non-convex problems. Empirical results on deep neural networks show that S-SNGD achieves better performance than SNGD in terms of both training loss and test accuracy.

The present study aimed to investigate the roles of cancer-associated fibroblasts (CAFs), matrix metalloproteinase-9 (MMP-9) and lymphatic vessel density (LVD) during the progression from adenocarcinoma in situ (AIS) to invasive lung adenocarcinoma (IAC). A total of 77 patients with stage 0-IA lung adenocarcinoma were enrolled. The expression levels of α-smooth muscle actin, MMP-9 and D2-40 were immunohistochemically analyzed. Survival analysis was performed using the Kaplan-Meier method. In the non-invasive component, the proportion of CAFs and the expression levels of MMP-9 increased from AIS to IAC; however, the LVD was not significantly different. CAFs were positively correlated with levels of MMP-9. The LVD had no significant correlation with CAFs and MMP-9. In the invasive component, CAFs, MMP-9 and LVD were significantly higher in IAC compared with in minimally invasive adenocarcinoma. CAFs, MMP-9 and LVD were all positively correlated with each other. The micropapillary subtype in IAC was associated with overall survival (OS). The LVD in IAC, but not MMP-9 and CAFs, was associated with OS. CAFs, MMP-9 and LVD were involved in the progression from AIS to IAC. CAFs exhibited a strong association with MMP-9 levels in the non-invasive and invasive components. The increase in the proportion of CAFs and the expression levels of MMP-9 may have been an early event before the adenocarcinoma became invasive. Once the adenocarcinoma was invasive, the LVD served an important role in tumor invasion and metastasis, and hence may be used as a prognostic marker of poor OS in stage IA IAC.

Semantic segmentation is an important task in medical image analysis. In general, training models with high performance needs a large amount of labeled data. However, collecting labeled data is typically difficult, especially for medical images. Several semi-supervised methods have been proposed to use unlabeled data to facilitate learning. Most of these methods use a self-training framework, in which the model cannot be well trained if the pseudo masks predicted by the model itself are of low quality. Co-training is another widely used semi-supervised method in medical image segmentation. It uses two models and makes them learn from each other. All these methods are not end-to-end. In this paper, we propose a novel end-to-end approach, called difference minimization network (DMNet), for semi-supervised semantic segmentation. To use unlabeled data, DMNet adopts two decoder branches and minimizes the difference between soft masks generated by the two decoders. In this manner, each decoder can learn under the supervision of the other decoder, thus they can be improved at the same time. Also, to make the model generalize better, we force the model to generate low-entropy masks on unlabeled data so the decision boundary of model lies in low-density regions. Meanwhile, adversarial training strategy is adopted to learn a discriminator which can encourage the model to generate more accurate masks. Experiments on a kidney tumor dataset and a brain tumor dataset show that our method can outperform the baselines, including both supervised and semi-supervised ones, to achieve the best performance.

Retrieving content relevant images from a large-scale fine-grained dataset could suffer from intolerably slow query speed and highly redundant storage cost, due to high-dimensional real-valued embeddings which aim to distinguish subtle visual differences of fine-grained objects. In this paper, we study the novel fine-grained hashing topic to generate compact binary codes for fine-grained images, leveraging the search and storage efficiency of hash learning to alleviate the aforementioned problems. Specifically, we propose a unified end-to-end trainable network, termed as ExchNet. Based on attention mechanisms and proposed attention constraints, ExchNet can firstly obtain both local and global features to represent object parts and the whole fine-grained objects, respectively. Furthermore, to ensure the discriminative ability and semantic meaning’s consistency of these part-level features across images, we design a local feature alignment approach by performing a feature exchanging operation. Later, an alternating learning algorithm is employed to optimize the whole ExchNet and then generate the final binary hash codes. Validated by extensive experiments, our ExchNet consistently outperforms state-of-the-art generic hashing methods on five fine-grained datasets. Moreover, compared with other approximate nearest neighbor methods, ExchNet achieves the best speed-up and storage reduction, revealing its efficiency and practicality.

With the explosive growth of video data in real applications, near-duplicate video retrieval (NDVR) has become indispensable and challenging, especially for short videos. However, all existing NDVR datasets are introduced for long videos. Furthermore, most of them are small-scale and lack of diversity due to the high cost of collecting and labeling near-duplicate videos. In this paper, we introduce a large-scale short video dataset, called SVD, for the NDVR task. SVD contains over 500,000 short videos and over 30,000 labeled videos of near-duplicates. We use multiple video mining techniques to construct positive/negative pairs. Furthermore, we design temporal and spatial transformations to mimic user-attack behavior in real applications for constructing more difficult variants of SVD. Experiments show that existing state-of-the-art NDVR methods, including real-value based and hashing based methods, fail to achieve satisfactory performance on this challenging dataset. The release of SVD dataset will foster research and system engineering in the NDVR area. The SVD dataset is available at https://svdbase.github.io.

End-to-end speaker embedding systems have shown promising performance on speaker verification tasks. Traditional end-to-end systems typically adopt softmax loss as training criterion, which is not strong enough for training discriminative models. In this paper, we adapt the additive margin softmax (AM-Softmax) loss, which is originally proposed for face verification, to speaker embedding systems. Furthermore, we propose a novel ensemble loss, called ensemble additive margin softmax (EAM-Softmax) loss, for speaker embedding by integrating Hilbert-Schmidt independence criterion (HSIC) into the speaker embedding system with the AM-Softmax loss. Experiments on a large-scale dataset VoxCeleb show that AM-Softmax loss is better than traditional loss functions, and approaches using EAM-Softmax loss can outperform existing speaker verification methods to achieve state-of-the-art performance.

Answer selection is an important subtask of question answering (QA), in which deep models usually achieve better performance than non-deep models. Most deep models adopt question-answer interaction mechanisms, such as attention, to get vector representations for answers. When these interaction based deep models are deployed for online prediction, the representations of all answers need to be recalculated for each question. This procedure is time-consuming for deep models with complex encoders like BERT which usually have better accuracy than simple encoders. One possible solution is to store the matrix representation (encoder output) of each answer in memory to avoid recalculation. But this will bring large memory cost. In this paper, we propose a novel method, called hashing based answer selection (HAS), to tackle this problem. HAS adopts a hashing strategy to learn a binary matrix representation for each answer, which can dramatically reduce the memory cost for storing the matrix representations of answers. Hence, HAS can adopt complex encoders like BERT in the model, but the online prediction of HAS is still fast with a low memory cost. Experimental results on three popular answer selection datasets show that HAS can outperform existing models to achieve state-of-the-art performance.

The present study aimed to investigate the roles of cancer-associated fibroblasts (CAFs), matrix metalloproteinase-9 (MMP-9) and lymphatic vessel density (LVD) during the progression from adenocarcinoma in situ (AIS) to invasive lung adenocarcinoma (IAC). A total of 77 patients with stage 0-IA lung adenocarcinoma were enrolled. The expression levels of α-smooth muscle actin, MMP-9 and D2-40 were immunohistochemically analyzed. Survival analysis was performed using the Kaplan-Meier method. In the non-invasive component, the proportion of CAFs and the expression levels of MMP-9 increased from AIS to IAC; however, the LVD was not significantly different. CAFs were positively correlated with levels of MMP-9. The LVD had no significant correlation with CAFs and MMP-9. In the invasive component, CAFs, MMP-9 and LVD were significantly higher in IAC compared with in minimally invasive adenocarcinoma. CAFs, MMP-9 and LVD were all positively correlated with each other. The micropapillary subtype in IAC was associated with overall survival (OS). The LVD in IAC, but not MMP-9 and CAFs, was associated with OS. CAFs, MMP-9 and LVD were involved in the progression from AIS to IAC. CAFs exhibited a strong association with MMP-9 levels in the non-invasive and invasive components. The increase in the proportion of CAFs and the expression levels of MMP-9 may have been an early event before the adenocarcinoma became invasive. Once the adenocarcinoma was invasive, the LVD served an important role in tumor invasion and metastasis, and hence may be used as a prognostic marker of poor OS in stage IA IAC.

Hashing has been widely used for large-scale search due to its low storage cost and fast query speed. By using supervised information, supervised hashing can significantly outperform unsupervised hashing. Recently, discrete supervised hashing and feature learning based deep hashing are two representative progresses in supervised hashing. On one hand, hashing is essentially a discrete optimization problem. Hence, utilizing supervised information to directly guide discrete (binary) coding procedure can avoid sub-optimal solution and improve the accuracy. On the other hand, feature learning based deep hashing, which integrates deep feature learning and hash-code learning into an end-to-end architecture, can enhance the feedback between feature learning and hash-code learning. The key in discrete supervised hashing is to adopt supervised information to directly guide the discrete coding procedure in hashing. The key in deep hashing is to adopt the supervised information to directly guide the deep feature learning procedure. However, most deep supervised hashing methods cannot use the supervised information to directly guide both discrete (binary) coding procedure and deep feature learning procedure in the same framework. In this paper, we propose a novel deep hashing method, called deep discrete supervised hashing (DDSH). DDSH is the first deep hashing method which can utilize pairwise supervised information to directly guide both discrete coding procedure and deep feature learning procedure and thus enhance the feedback between these two important procedures. Experiments on four real datasets show that DDSH can outperform other state-of-the-art baselines, including both discrete hashing and deep hashing baselines, for image retrieval.

Due to its successful application in recommender systems, collaborative filtering (CF) has become a hot research topic in data mining and information retrieval. In traditional CF methods, only the feedback matrix, which contains either explicit feedback (also called ratings) or implicit feedback on the items given by users, is used for training and prediction. Typically, the feedback matrix is sparse, which means that most users interact with few items. Due to this sparsity problem, traditional CF with only feedback information will suffer from unsatisfactory performance. Recently, many researchers have proposed to utilize auxiliary information, such as item content (attributes), to alleviate the data sparsity problem in CF. Collaborative topic regression (CTR) is one of these methods which has achieved promising performance by successfully integrating both feedback information and item content information. In many real applications, besides the feedback and item content information, there may exist relations (also known as networks) among the items which can be helpful for recommendation. In this paper, we develop a novel hierarchical Bayesian model called Relational Collaborative Topic Regression (RCTR), which extends CTR by seamlessly integrating the user-item feedback information, item content information, and network structure among items into the same model. Experiments on real-world datasets show that our model can achieve better prediction accuracy than the state-of-the-art methods with lower empirical training time. Moreover, RCTR can learn good interpretable latent structures which are useful for recommendation.

Generalization of large margin classification methods from the binary classification setting to the more general multicategory setting is often found to be non-trivial. In this paper, we study large margin classification methods that can be seamlessly applied to both settings, with the binary setting simply as a special case. In particular, we explore the Fisher consistency properties of multicategory majorization losses and present a construction framework of majorization losses of the 0-1 loss. Under this framework, we conduct an in-depth analysis about three widely used multicategory hinge losses. Corresponding to the three hinge losses, we propose three multicategory majorization losses based on a coherence function. The limits of the three coherence losses as the temperature approaches zero are the corresponding hinge losses, and the limits of the minimizers of their expected errors are the minimizers of the expected errors of the corresponding hinge losses. Finally, we develop multicategory large margin classification methods by using a so-called multiclass C-loss.

Recently, deep learning has been successfully applied to the problem of hashing, yielding remarkable performance compared to traditional methods with hand-crafted features. However, most of existing deep hashing methods are designed for the supervised scenario and require a large number of labeled data. In this paper, we propose a novel semi-supervised hashing method for image retrieval, named Deep Hashing with a Bipartite Graph (DHBG), to simultaneously learn embeddings, features and hash codes. More specifically, we construct a bipartite graph to discover the underlying structure of data, based on which an embedding is generated for each instance. Then, we feed raw pixels as well as embeddings to a deep neural network, and concatenate the resulting features to determine the hash code. Compared to existing methods, DHBG is a universal framework that is able to utilize various types of graphs and losses. Furthermore, we propose an inductive variant of DHBG to support out-of-sample extensions. Experimental results on real datasets show that our DHBG outperforms state-of-the-art hashing methods.

Stochastic gradient descent (SGD) and its variants have become more and more popular in machine learning due to their efficiency and effectiveness. To handle large-scale problems, researchers have recently proposed several parallel SGD methods for multicore systems. However, existing parallel SGD methods cannot achieve satisfactory performance in real applications. In this paper, we propose a fast asynchronous parallel SGD method, called AsySVRG, by designing an asynchronous strategy to parallelize the recently proposed SGD variant called stochastic variance reduced gradient (SVRG). AsySVRG adopts a lock-free strategy which is more efficient than other strategies with locks. Furthermore, we theoretically prove that AsySVRG is convergent with a linear convergence rate. Both theoretical and empirical results show that AsySVRG can outperform existing state-of-the-art parallel SGD methods like Hogwild! in terms of convergence rate and computation cost.

By leveraging semantic (label) information, supervised hashing has demonstrated better accuracy than unsupervised hashing in many real applications. Because the hashing-code learning problem is essentially a discrete optimization problem which is hard to solve, most existing supervised hashing methods try to solve a relaxed continuous optimization problem by dropping the discrete constraints. However, these methods typically suffer from poor performance due to the errors caused by the relaxation. Some other methods try to directly solve the discrete optimization problem. However, they are typically time-consuming and unscalable. In this paper, we propose a novel method, called column sampling based discrete supervised hashing (COSDISH), to directly learn the discrete hashing code from semantic information. COSDISH is an iterative method, in each iteration of which several columns are sampled from the semantic similarity matrix and then the hashing code is decomposed into two parts which can be alternately optimized in a discrete way. Theoretical analysis shows that the learning (optimization) algorithm of COSDISH has a constant-approximation bound in each step of the alternating optimization procedure. Empirical results on datasets with semantic labels illustrate that COSDISH can outperform the state-of-the-art methods in real applications like image retrieval.

Effectiveness of epoxy resin filled microcapsules was investigated for healing of cracks generated in coatings. Microcapsules were prepared by in situ polymerization of urea–formaldehyde resin to form shell over epoxy resin droplets. Characteristics of these capsules were studied by 3D measuring laser microscope, particle size analyzer, Fourier-transform infrared spectroscopy (FTIR) and differential scanning calorimeter (DSC) to investigate their surface morphology, size distribution, chemical structure and thermal stability, respectively. The results indicate that microcapsules containing epoxy resins can be synthesized successfully. The size is around 100 μm. The rough outer surface of microcapsule is composed of agglomerated urea–formaldehyde nanoparticles. The size and surface morphology of microcapsule can be controlled by selecting different processing parameters. The microcapsules basically exhibit good storage stability at room temperature, and they are chemically stable before the heating temperature is up to approximately 200 °C. The model system of self-healing coating consists of epoxy resin matrix, 10 wt% microencapsulated healing agent, 2 wt% catalyst solution. The self-healing function of this coating system is evaluated through self-healing testing of damaged and healed coated steel samples.

We propose a novel probabilistic model for collaborative filtering (CF), called SRMCoFi, which seamlessly integrates both linear and bilinear random effects into a principled framework. The formulation of SRMCoFi is supported by both social psychological experiments and statistical theories. Not only can many existing CF methods be seen as special cases of SRMCoFi, but it also integrates their advantages while simultaneously overcoming their disadvantages. The solid theoretical foundation of SRMCoFi is further supported by promising empirical results obtained in extensive experiments using real CF data sets on movie ratings.

Microencapsulated healing agents that possess adequate strength, long shelf-life, and excellent bonding to the host material are required for self-healing materials. The in situ encapsulation method is demonstrated over an order of magnitude size reduction for the preparation of urea-formaldehyde (UF) capsules filled with a healing agent, a mixture epoxy resin of the epoxy 711 and E-51. Capsules with diameters as small as about 100 μm are achieved under the agitation rate of 800 r min−1. The capsules possess a uniform UF shell wall (4 μm average thickness). By using the analysis of scanning electronic microscope (SEM), thermal analysis (TG-DTA) and FTIR, the characteristics of the microcapsules were investigated respectively. Successful self-healing has been demonstrated for anticorrosion covering materials with microcapsules.

With the development of the embedded microcapsule concept for self-healing material, the preparation of microcapsule has been paid more attentions. A new series of microcapsules were prepared by in situ polymerization technology in an oil-in-water emulsion with polyoxymethylene urea (PMU) as shell material and a mixture of epoxy resins as core material. The PMU microcapsules were characterized by Fourier transform infrared spectroscopy (FTIR), scanning electronic microscopy (SEM), particle size analyzer and thermo gravimetric analyzer (TGA) to investigate their chemical structure, surface morphology, size distribution and thermal stability, respectively. The results indicate that PMU microcapsules containing epoxy resins can be synthesized successfully. The optimized reaction parameters were obtained as follow: agitation rate 600 rpm, 60°C water bath, pH=3.5, core material 20ml and hot water dilution by in-situ polymerization. The size is around 116 μm. The rough outer surface of microcapsule is composed of agglomerated PMU nanoparticles. The microcapsules basically exhibit good storage stability at room temperature, and they are chemically stable before the heating temperature is up to approximately 200°C.

In this paper, we propose a novel supervised extension of GPLVM, called Gaussian process latent random field (GPLRF), by enforcing the latent variables to be a Gaussian Markov random field with respect to a graph constructed from the supervisory information.

Matrix factorization (MF) has become the most popular technique for recommender systems due to its promising performance. Recently, distributed (parallel) MF models have received much attention from researchers of big data community. In this paper, we propose a novel model, called distributed stochastic alternating direction methods of multipliers (DS-ADMM), for large-scale MF problems. DS-ADMM is a distributed stochastic variant of ADMM. In particular, we first devise a new data split strategy to make the distributed MF problem fit for the ADMM framework. Then, a stochastic ADMM scheme is designed for learning. Finally, we implement DS-ADMM based on message passing interface (MPI), which can run on clusters with multiple machines (nodes). Experiments on several data sets from recommendation applications show that our DS-ADMM model can outperform other state-of-the-art distributed MF models in terms of both efficiency and accuracy.

Due to its low storage cost and fast query speed, hashing has been widely adopted for similarity search in multimedia data. In particular, more and more attentions have been payed to multimodal hashing for search in multimedia data with multiple modalities, such as images with tags. Typically, supervised information of semantic labels is also available for the data points in many real applications. Hence, many supervised multimodal hashing~(SMH) methods have been proposed to utilize such semantic labels to further improve the search accuracy. However, the training time complexity of most existing SMH methods is too high, which makes them unscalable to large-scale datasets. In this paper, a novel SMH method, called semantic correlation maximization~(SCM), is proposed to seamlessly integrate semantic labels into the hashing learning procedure for large-scale data modeling. Experimental results on two real-world datasets show that SCM can significantly outperform the state-of-the-art SMH methods, in terms of both accuracy and scalability.

In general, a large amount of labels are needed for supervised learning algorithms to achieve satisfactory performance. It's typically very time-consuming and money-consuming to get such kind of labeled data. Recently, crowdsourcing services provide an effective way to collect labeled data with much lower cost. Hence, crowdsourced learning (CL), which performs learning with labeled data collected from crowdsourcing services, has become a very hot and interesting research topic in recent years. Most existing CL methods exploit only the labels from different workers (annotators) for learning while ignoring the attributes of the instances. In many real applications, the attributes of the instances are actually the most discriminative information for learning. Hence, CL methods with attributes have attracted more and more attention from CL researchers. One representative model of such kind is the personal classifier (PC) model, which has achieved the state-of-the-art performance. However, the PC model makes an unreasonable assumption that all the workers contribute equally to the final classification. This contradicts the fact that different workers have different quality (ability) for data labeling. In this paper, we propose a novel model, called robust personal classifier (RPC), for robust crowdsourced learning. Our model can automatically learn an expertise score for each worker. This expertise score reflects the inherent quality of each worker. The final classifier of our RPC model gives high weights for good workers and low weights for poor workers or spammers, which is more reasonable than PC model with equal weights for all workers. Furthermore, the learned expertise score can be used to eliminate spammers or low-quality workers. Experiments on simulated datasets and UCI datasets show that the proposed model can dramatically outperform the baseline models such as PC model in terms of classification accuracy and ability to detect spammers.

Hashing is used to learn binary-code representation for data with expectation of preserving the neighborhood structure in the original feature space. Due to its fast query speed and reduced storage cost, hashing has been widely used for efficient nearest neighbor search in a large variety of applications like text and image retrieval. Most existing hashing methods adopt Hamming distance to measure the similarity (neighborhood) between points in the hashcode space. However, one problem with Hamming distance is that it may destroy the neighborhood structure in the original feature space, which violates the essential goal of hashing. In this paper, Manhattan hashing (MH), which is based on Manhattan distance, is proposed to solve the problem of Hamming distance based hashing. The basic idea of MH is to encode each projected dimension with multiple bits of natural binary code (NBC), based on which the Manhattan distance between points in the hashcode space is calculated for nearest neighbor search. MH can effectively preserve the neighborhood structure in the data to achieve the goal of hashing. To the best of our knowledge, this is the first work to adopt Manhattan distance with NBC for hashing. Experiments on several large-scale image data sets containing up to one million points show that our MH method can significantly outperform other state-of-the-art methods.

Besides the rating information, an increasing number of modern recommender systems also allow the users to add personalized tags to the items. Such tagging information may provide very useful information for item recommendation, because the users' interests in items can be implicitly reflected by the tags that they often use. Although some content-based recommender systems have made preliminary attempts recently to utilize tagging information to improve the recommendation performance, few recommender systems based on collaborative filtering (CF) have employed tagging information to help the item recommendation procedure. In this paper, we propose a novel framework, called tag informed collaborative filtering (TagiCoFi), to seamlessly integrate tagging information into the CF procedure. Experimental results demonstrate that TagiCoFi outperforms its counterpart which discards the tagging information even when it is available, and achieves state-of-the-art performance.

Over the past decade, multiple-instance learning (MIL) has been successfully utilized to model the localized content-based image retrieval (CBIR) problem, in which a bag corresponds to an image and an instance corresponds to a region in the image. However, existing feature representation schemes are not effective enough to describe the bags in MIL, which hinders the adaptation of sophisticated single-instance learning (SIL) methods for MIL problems. In this paper, we first propose an evidence region (or evidence instance) identification method to identify the evidence regions supporting the labels of the images (i.e., bags). Then, based on the identified evidence regions, a very effective feature representation scheme, which is also very computationally efficient and robust to labeling noise, is proposed to describe the bags. As a result, the MIL problem is converted into a standard SIL problem and a support vector machine (SVM) can be easily adapted for localized CBIR. Experimental results on two challenging data sets show that our method, called EC-SVM, can outperform the state-of-the-art methods in terms of accuracy, robustness and efficiency.

A fundamental challenge in face recognition lies in determining what facial features are important for the identification of faces. In this paper, a novel face recognition framework is proposed to address this problem. In our framework, 3D face models are used to synthesize a huge database of realistic face images which covers wide appearance variations of faces due to various pose, illumination, and expression changes. A novel feature selection algorithm which we call Joint Boosting is developed to extract discriminative face features using this massive database. The major contributions of this paper are: (1) With the help of 3D face models, a massive database of realistic virtual face images is generated to achieve robust feature selection; (2)Because the huge database covers a wide range of face variations, our feature selection procedure only needs to be trained once, and the selected feature set can be generalized to other face database without re-training; (3) We propose a new learning algorithm, Joint Boosting Algorithm, which is effective and efficient in learning directly from a massive database without having to convert face images to intra-personal and extra-personal difference images. This property is important for applying our algorithm to other general pattern recognition problems. Experimental results show that our method significantly improves recognition performance.

An illumination invariant face recognition method based on neural network ensemble architecture is proposed. Given a face image with an arbitrary illumination direction, it can complete recognition in a uniform way with high performance without knowing or estimating the illumination direction. Experimental result shows that the recognition ratio of the ensemble architecture is higher than the conventional approach that uses a single neural network to recognize faces of a specific illumination direction.

study laser. But we think that it needs revise too.The photon degeneracy of laser beam is the most nature of lasers. macroscopic parameter is power (energy)and coherence. In there. coherence includes the space coherence and time coherence. if we discuss the problem ofspace quality oflaser beam we should discuss the beam space coherence except for power and energy. It is luckly.space coherence is determind by the product of beam waist spot—size and divergence angle of far field and haverelation with the space intensity distribution of beam.