arXiv subjects

Yan Wang

Publications and source records attributed to Yan Wang.

At least 955 records · Page 53Linked to original sources

arXiv2019

Using Machine Learning and Natural Language Processing to Review and Classify the Medical Literature on Cancer Susceptibility Genes

PURPOSE: The medical literature relevant to germline genetics is growing exponentially. Clinicians need tools monitoring and prioritizing the literature to understand the clinical implications of the pathogenic genetic variants. We developed and evaluated two machine learning models to classify abstracts as relevant to the penetrance (risk of cancer for germline mutation carriers) or prevalence of germline genetic mutations. METHODS: We conducted literature searches in PubMed and retrieved paper titles and abstracts to create an annotated dataset for training and evaluating the two machine learning classification models. Our first model is a support vector machine (SVM) which learns a linear decision rule based on the bag-of-ngrams representation of each title and abstract. Our second model is a convolutional neural network (CNN) which learns a complex nonlinear decision rule based on the raw title and abstract. We evaluated the performance of the two models on the classification of papers as relevant to penetrance or prevalence. RESULTS: For penetrance classification, we annotated 3740 paper titles and abstracts and used 60% for training the model, 20% for tuning the model, and 20% for evaluating the model. The SVM model achieves 89.53% accuracy (percentage of papers that were correctly classified) while the CNN model achieves 88.95 % accuracy. For prevalence classification, we annotated 3753 paper titles and abstracts. The SVM model achieves 89.14% accuracy while the CNN model achieves 89.13 % accuracy. CONCLUSION: Our models achieve high accuracy in classifying abstracts as relevant to penetrance or prevalence. By facilitating literature review, this tool could help clinicians and researchers keep abreast of the burgeoning knowledge of gene-cancer associations and keep the knowledge bases for clinical decision support tools up to date.

cs.IR↗

arXiv2019

Predicting class-imbalanced business risk using resampling, regularization, and model ensembling algorithms

We aim at developing and improving the imbalanced business risk modeling via jointly using proper evaluation criteria, resampling, cross-validation, classifier regularization, and ensembling techniques. Area Under the Receiver Operating Characteristic Curve (AUC of ROC) is used for model comparison based on 10-fold cross validation. Two undersampling strategies including random undersampling (RUS) and cluster centroid undersampling (CCUS), as well as two oversampling methods including random oversampling (ROS) and Synthetic Minority Oversampling Technique (SMOTE), are applied. Three highly interpretable classifiers, including logistic regression without regularization (LR), L1-regularized LR (L1LR), and decision tree (DT) are implemented. Two ensembling techniques, including Bagging and Boosting, are applied on the DT classifier for further model improvement. The results show that, Boosting on DT by using the oversampled data containing 50% positives via SMOTE is the optimal model and it can achieve AUC, recall, and F1 score valued 0.8633, 0.9260, and 0.8907, respectively.

stat.ML↗

arXiv2019

The parameters of a family of linear codes

A large family of linear codes with flexible parameters from almost bent functions and perfect nonlinear functions are constructed and their parameters are determined. Some constructed linear codes and their related codes are optimal in the sense that they meet certain bounds on linear codes. Applications of these codes in secret sharing schemes and combinational designs are presented.

cs.IT↗

arXiv2019

Anytime Stereo Image Depth Estimation on Mobile Devices

Many applications of stereo depth estimation in robotics require the generation of accurate disparity maps in real time under significant computational constraints. Current state-of-the-art algorithms force a choice between either generating accurate mappings at a slow pace, or quickly generating inaccurate ones, and additionally these methods typically require far too many parameters to be usable on power- or memory-constrained devices. Motivated by these shortcomings, we propose a novel approach for disparity prediction in the anytime setting. In contrast to prior work, our end-to-end learned approach can trade off computation and accuracy at inference time. Depth estimation is performed in stages, during which the model can be queried at any time to output its current best estimate. Our final model can process 1242$ \times $375 resolution images within a range of 10-35 FPS on an NVIDIA Jetson TX2 module with only marginal increases in error -- using two orders of magnitude fewer parameters than the most competitive baseline. The source code is available at https://github.com/mileyan/AnyNet .

cs.CV↗

arXiv2019

Search for Extra Scalars Produced in Association with Muon Pairs at the ILC

We study the search for an extra scalar $S$ boson produced in association with the $Z$ boson at the International Linear Collider (ILC). The study is performed at center-of-mass energies of 250 GeV and 500 GeV based on the full simulation of the International Large Detector (ILD). In order to be as model-independent as possible, the analysis uses the recoil technique, in particular with the $Z$ boson decaying into a pair of muons. As a result, exclusion cross-section limits are given in terms of a scale factor $k$ with respect to the Standard Model Higgs-strahlung process cross section. These predicted results, covering all possible searching regions of the extra scalars at the 250 GeV ILC and the 500 GeV ILC, can be interpreted independently of the decay modes of the $S$ boson.

hep-ex↗

arXiv2019

Enhancement of coherent dipole coupling between two atoms via squeezing a cavity mode

We propose a theoretical method to enhance the coherent dipole coupling between two atoms in an optical cavity via parametrically squeezing the cavity mode. In the present scheme, conditions for coherent coupling are derived in detail and diverse dynamics of the system can be obtained by regulating system parameters. In the presence of environmental noise, an auxiliary squeezed field is employed to suppress, and even completely eliminate the additional noise induced by squeezing. In addition, we demonstrate that our scheme enables the effective suppression of atomic spontaneous emission. The results in our investigation could be used for diverse applications in quantum technologies.

quant-ph↗

arXiv2019

Atomic-scale factors that control the rate capability of nanostructured amorphous Si for high-energy-density batteries

Nanostructured Si is the most promising high-capacity anode material to substantially increase the energy density of Li-ion batteries. Among the remaining challenges is its low rate capability as compared to conventional materials. To understand better what controls the diffusion of Li in the amorphous Li-Si alloy, we use a novel machine-learning potential trained on more than 40,000 ab-initio calculations and nanosecond-scale molecular dynamics simulations, to visualize for the first time the delithiation of entire LiSi nanoparticles. Our results show that the Si host is not static but undergoes a dynamic rearrangement from isolated atoms, to chains, and clusters, with the Li diffusion strongly governed by this Si rearrangement. We find that the Li diffusivity is highest when Si segregates into clusters, so that Li diffusion proceeds via hopping between the Si clusters. The average size of Si clusters and the concentration range over which Si clustering occurs can thus function as design criteria for the development of rate-improved anodes based on modified Si.

cond-mat.dis-nn↗

arXiv2019

A XGBoost risk model via feature selection and Bayesian hyper-parameter optimization

This paper aims to explore models based on the extreme gradient boosting (XGBoost) approach for business risk classification. Feature selection (FS) algorithms and hyper-parameter optimizations are simultaneously considered during model training. The five most commonly used FS methods including weight by Gini, weight by Chi-square, hierarchical variable clustering, weight by correlation, and weight by information are applied to alleviate the effect of redundant features. Two hyper-parameter optimization approaches, random search (RS) and Bayesian tree-structured Parzen Estimator (TPE), are applied in XGBoost. The effect of different FS and hyper-parameter optimization methods on the model performance are investigated by the Wilcoxon Signed Rank Test. The performance of XGBoost is compared to the traditionally utilized logistic regression (LR) model in terms of classification accuracy, area under the curve (AUC), recall, and F1 score obtained from the 10-fold cross validation. Results show that hierarchical clustering is the optimal FS method for LR while weight by Chi-square achieves the best performance in XG-Boost. Both TPE and RS optimization in XGBoost outperform LR significantly. TPE optimization shows a superiority over RS since it results in a significantly higher accuracy and a marginally higher AUC, recall and F1 score. Furthermore, XGBoost with TPE tuning shows a lower variability than the RS method. Finally, the ranking of feature importance based on XGBoost enhances the model interpretation. Therefore, XGBoost with Bayesian TPE hyper-parameter optimization serves as an operative while powerful approach for business risk modeling.

stat.ML↗

arXiv2019

An Automatic Interaction Detection Hybrid Model for Bankcard Response Classification

In this paper, we propose a hybrid bankcard response model, which integrates decision tree based chi-square automatic interaction detection (CHAID) into logistic regression. In the first stage of the hybrid model, CHAID analysis is used to detect the possibly potential variable interactions. Then in the second stage, these potential interactions are served as the additional input variables in logistic regression. The motivation of the proposed hybrid model is that adding variable interactions may improve the performance of logistic regression. To demonstrate the effectiveness of the proposed hybrid model, it is evaluated on a real credit customer response data set. As the results reveal, by identifying potential interactions among independent variables, the proposed hybrid approach outperforms the logistic regression without searching for interactions in terms of classification accuracy, the area under the receiver operating characteristic curve (ROC), and Kolmogorov-Smirnov (KS) statistics. Furthermore, CHAID analysis for interaction detection is much more computationally efficient than the stepwise search mentioned above and some identified interactions are shown to have statistically significant predictive power on the target variable. Last but not least, the customer profile created based on the CHAID tree provides a reasonable interpretation of the interactions, which is the required by regulations of the credit industry. Hence, this study provides an alternative for handling bankcard classification tasks.

stat.ML↗

arXiv2018

Reducing the Upfront Cost of Private Clouds with Clairvoyant Virtual Machine Placement

Although public clouds still occupy the largest portion of the total cloud infrastructure, private clouds are attracting increasing interest from both industry and academia because of their better security and privacy control. According to the existing studies, the high upfront cost is among the most critical challenges associated with private clouds. To reduce cost and improve performance, virtual machine placement (VMP) methods have been extensively investigated, however, few of these methods have focused on private clouds. This paper proposes a heterogeneous and multidimensional clairvoyant dynamic bin packing (CDBP) model, in which the scheduler can conduct more efficient VMP processes using additional information on the arrival time and duration of virtual machines to reduce the datacenter scale and thereby decrease the upfront cost of private clouds. In addition, a novel branch-and-bound algorithm with a divide-and-conquer strategy (DCBB) is proposed to effectively and efficiently handle the derived problem. One state-of-the-art and several classic VMP methods are also modified to adapt to the proposed model to observe their performance and compare with our proposed algorithm. Extensive experiments are conducted on both real-world and synthetic workloads to evaluate the accuracy and efficiency of the algorithms. The experimental results demonstrate that DCBB delivers near-optimal solutions with a convergence rate that is much faster than those of the other search-based algorithms evaluated. In particular, DCBB yields the optimal solution for a real-world workload with an execution time that is an order of magnitude shorter than that required by the original branch-and-bound (BB) algorithm.

cs.DC↗

arXiv2018

A two-stage hybrid model by using artificial neural networks as feature construction algorithms

We propose a two-stage hybrid approach with neural networks as the new feature construction algorithms for bankcard response classifications. The hybrid model uses a very simple neural network structure as the new feature construction tool in the first stage, then the newly created features are used as the additional input variables in logistic regression in the second stage. The model is compared with the traditional one-stage model in credit customer response classification. It is observed that the proposed two-stage model outperforms the one-stage model in terms of accuracy, the area under ROC curve, and KS statistic. By creating new features with the neural network technique, the underlying nonlinear relationships between variables are identified. Furthermore, by using a very simple neural network structure, the model could overcome the drawbacks of neural networks in terms of its long training time, complex topology, and limited interpretability.

stat.ML↗

arXiv2018

Robust Face Detection via Learning Small Faces on Hard Images

Recent anchor-based deep face detectors have achieved promising performance, but they are still struggling to detect hard faces, such as small, blurred and partially occluded faces. A reason is that they treat all images and faces equally, without putting more effort on hard ones; however, many training images only contain easy faces, which are less helpful to achieve better performance on hard images. In this paper, we propose that the robustness of a face detector against hard faces can be improved by learning small faces on hard images. Our intuitions are (1) hard images are the images which contain at least one hard face, thus they facilitate training robust face detectors; (2) most hard faces are small faces and other types of hard faces can be easily converted to small faces by shrinking. We build an anchor-based deep face detector, which only output a single feature map with small anchors, to specifically learn small faces and train it by a novel hard image mining strategy. Extensive experiments have been conducted on WIDER FACE, FDDB, Pascal Faces, and AFW datasets to show the effectiveness of our method. Our method achieves APs of 95.7, 94.9 and 89.7 on easy, medium and hard WIDER FACE val dataset respectively, which surpass the previous state-of-the-arts, especially on the hard subset. Code and model are available at https://github.com/bairdzhang/smallhardface.

cs.CV↗

arXiv2018

High contrast linking 6 lasers to a 1 GHz Ybfiber laser frequency comb

We demonstrate a 0.95 GHz repetition rate fully stabilized Yb:fiber frequency comb without optical amplification. Benefitted from the high mode power and high coherence, this comb achieved 35 dB to 42 dB signal to noise ratio on the direct heterodyne beat signals with at least 6 continuous wave lasers (at 580nm, 679nm, 698nm, 707nm, 813nm and 922 nm) while keeping >40 dB carrier envelop frequency signal. It can be used for the direct measurement of optical frequencies in visible and near-infrared wavelengths, and has a great potential on simultaneous comparison of multiple optical frequencies.

physics.optics↗

arXiv2018

Probing the Higgs boson-gluon coupling via the jet energy profile at $e^+e^-$ colliders

The effective coupling of the Higgs boson to a gluon pair is one of the most important parameters to test the Standard Model and search for the new physics beyond. In this paper, we propose several new observables based on the jet energy profile to extract the effective coupling. The statistical uncertainties of the effective coupling extracted by using new observables are derived and estimated based on the simulation at the future $e^+e^-$ collider for $250$ GeV center-of-mass energy and 5 ab$^{-1}$ integrated luminosity. We found that the statistical uncertainties of effective coupling via the optimized observable can reach about $1.6\%$ in the channels of a $Z$ boson decaying to lepton pairs and is reduced by $52\%$ compared to the relevant uncertainties in the conventional approach. These new observables potentially can be helpful for the measurement of effective coupling at future $e^+e^-$ colliders.

hep-ph↗

arXiv2018

Semi-Supervised Multi-Organ Segmentation via Deep Multi-Planar Co-Training

In multi-organ segmentation of abdominal CT scans, most existing fully supervised deep learning algorithms require lots of voxel-wise annotations, which are usually difficult, expensive, and slow to obtain. In comparison, massive unlabeled 3D CT volumes are usually easily accessible. Current mainstream works to address the semi-supervised biomedical image segmentation problem are mostly graph-based. By contrast, deep network based semi-supervised learning methods have not drawn much attention in this field. In this work, we propose Deep Multi-Planar Co-Training (DMPCT), whose contributions can be divided into two folds: 1) The deep model is learned in a co-training style which can mine consensus information from multiple planes like the sagittal, coronal, and axial planes; 2) Multi-planar fusion is applied to generate more reliable pseudo-labels, which alleviates the errors occurring in the pseudo-labels and thus can help to train better segmentation networks. Experiments are done on our newly collected large dataset with 100 unlabeled cases as well as 210 labeled cases where 16 anatomical structures are manually annotated by four radiologists and confirmed by a senior expert. The results suggest that DMPCT significantly outperforms the fully supervised method by more than 4% especially when only a small set of annotations is used.

cs.CV↗

arXiv2018

Translating a Math Word Problem to an Expression Tree

Sequence-to-sequence (SEQ2SEQ) models have been successfully applied to automatic math word problem solving. Despite its simplicity, a drawback still remains: a math word problem can be correctly solved by more than one equations. This non-deterministic transduction harms the performance of maximum likelihood estimation. In this paper, by considering the uniqueness of expression tree, we propose an equation normalization method to normalize the duplicated equations. Moreover, we analyze the performance of three popular SEQ2SEQ models on the math word problem solving. We find that each model has its own specialty in solving problems, consequently an ensemble model is then proposed to combine their advantages. Experiments on dataset Math23K show that the ensemble model with equation normalization significantly outperforms the previous state-of-the-art methods.

cs.CL↗

arXiv2018

Temperature-filling phase diagram of the two-dimensional Holstein model in the thermodynamic limit by self-consistent Migdal approximation

We study the temperature-filling phase diagram of the single-band Holstein model in two dimensions using the self-consistent Migdal approximation, where both the electron and phonon self-energies are treated on an equal footing. By employing an efficient numerical algorithm utilizing fast Fourier transforms to evaluate momentum and Matsubara frequency summations, we determine the charge-density-wave (CDW) and superconducting transition temperatures in the thermodynamic limit using lattice sizes that are sufficient to eliminate significant finite size effects present at lower temperatures. We obtain the temperature-filling phase diagrams for a range of coupling strengths and phonon frequencies for the model defined on a square lattice with and without next-nearest neighbor hopping. We find the appearance of a superconducting dome with a critical temperature that decreases before reaching the $\mathbf{q}_{\text{max}} = (π,π)$ CDW phase boundary. For very low phonon frequencies, we also find an incommensurate CDW phase with the ordering vector $\mathbf{q}_{\text{max}} \approx (π,π)$ appearing between the commensurate CDW and superconducting phases. Our numerical implementation can be easily extended to treat momentum-dependent electron-phonon coupling, as well as dispersive phonon branches, and has been made available to the public.

cond-mat.supr-con↗

arXiv2018

Deep Person Re-identification for Probabilistic Data Association in Multiple Pedestrian Tracking

We present a data association method for vision-based multiple pedestrian tracking, using deep convolutional features to distinguish between different people based on their appearances. These re-identification (re-ID) features are learned such that they are invariant to transformations such as rotation, translation, and changes in the background, allowing consistent identification of a pedestrian moving through a scene. We incorporate re-ID features into a general data association likelihood model for multiple person tracking, experimentally validate this model by using it to perform tracking in two evaluation video sequences, and examine the performance improvements gained as compared to several baseline approaches. Our results demonstrate that using deep person re-ID for data association greatly improves tracking robustness to challenges such as occlusions and path crossings.

cs.CV↗