arXiv subjects

Ping Zhang

Publications and source records attributed to Ping Zhang.

At least 415 records · Page 23Linked to original sources

arXiv2020

When deep learning meets causal inference: a computational framework for drug repurposing from real-world data

Drug repurposing is an effective strategy to identify new uses for existing drugs, providing the quickest possible transition from bench to bedside. Existing methods for drug repurposing that mainly focus on pre-clinical information may exist translational issues when applied to human beings. Real world data (RWD), such as electronic health records and insurance claims, provide information on large cohorts of users for many drugs. Here we present an efficient and easily-customized framework for generating and testing multiple candidates for drug repurposing using a retrospective analysis of RWDs. Building upon well-established causal inference and deep learning methods, our framework emulates randomized clinical trials for drugs present in a large-scale medical claims database. We demonstrate our framework in a case study of coronary artery disease (CAD) by evaluating the effect of 55 repurposing drug candidates on various disease outcomes. We achieve 6 drug candidates that significantly improve the CAD outcomes but not have been indicated for treating CAD, paving the way for drug repurposing.

stat.AP↗

arXiv2020

DrugDBEmbed : Semantic Queries on Relational Database using Supervised Column Encodings

Traditional relational databases contain a lot of latent semantic information that have largely remained untapped due to the difficulty involved in automatically extracting such information. Recent works have proposed unsupervised machine learning approaches to extract such hidden information by textifying the database columns and then projecting the text tokens onto a fixed dimensional semantic vector space. However, in certain databases, task-specific class labels may be available, which unsupervised approaches are unable to lever in a principled manner. Also, when embeddings are generated at individual token level, then column encoding of multi-token text column has to be computed by taking the average of the vectors of the tokens present in that column for any given row. Such averaging approach may not produce the best semantic vector representation of the multi-token text column, as observed while encoding paragraphs or documents in natural language processing domain. With these shortcomings in mind, we propose a supervised machine learning approach using a Bi-LSTM based sequence encoder to directly generate column encodings for multi-token text columns of the DrugBank database, which contains gold standard drug-drug interaction (DDI) labels. Our text data driven encoding approach achieves very high Accuracy on the supervised DDI prediction task for some columns and we use those supervised column encodings to simulate and evaluate the Analogy SQL queries on relational data to demonstrate the efficacy of our technique.

cs.DB↗

arXiv2020

Global small analytic solutions of MHD boundary layer equations

In this paper, we prove the global existence and the large time decay estimate of solutions to the two-dimensional MHD boundary layer equations with small initial data, which is analytical in the tangential variable. The main idea of the proof is motivated by that of \cite{PZ5}. The additional difficulties are: 1. there appears the magnetic field; 2. the far field here depends on the tangential variable; 3. the Reynolds number is different from magnetic Reynolds number. In particular, we solved an open question in \cite{XY19} concerning the large time existence of analytical solutions to the MHD boundary layer equations.

math.AP↗

arXiv2020

Quantile regression for compositional covariates

Quantile regression is a very important tool to explore the relationship between the response variable and its covariates. Motivated by mean regression with LASSO for compositional covariates proposed by Lin et al. (2014), we consider quantile regression with no-penalty and penalty function. We develop the computational algorithms based on linear programming. Numerical studies indicate that our methods provides the better alternative than mean regression under many settings, particularly for heavy-tailed or skewed distribution of the error term. Finally, we study the fat data using the proposed method.

stat.ME↗

arXiv2020

Stability of Couette flow for 2D Boussinesq system with vertical dissipation

This paper establishes the nonlinear stability of the Couette flow for the 2D Boussinesq equations with only vertical dissipation. The Boussinesq equations concerned here model buoyancy-driven fluids such as atmospheric and oceanographic flows. Due to the presence of the buoyancy forcing, the energy of the standard Boussinesq equations could grow in time. It is the enhanced dissipation created by the linear non-self-adjoint operator $y\partial_x -ν\partial_{yy}$ in the perturbation equation that makes the nonlinear stability possible. When the initial perturbation from the Couette flow $(y, 0)$ is no more than the viscosity to a suitable power (in the Sobolev space $H^b$ with $b>\frac43$), we prove that the solution of the 2D Boussnesq system with only vertical dissipation on $\mathbb T\times \mathbb R$ remains close to the Couette at the same order. A special consequence of this result is the stability of the Couette for the 2D Navier-Stokes equations with only vertical dissipation.

math.AP↗

arXiv2020

On the radius of analyticity of solutions to semi-linear parabolic systems

We study the radius of analyticity~$R(t)$ in space, of strong solutions to systems of scale-invariant semi-linear parabolic equations. It is well-known that near the initial time,~$R(t)t^{-\frac12}$ is bounded from below by a positive constant. In this paper we prove that~$\displaystyle\liminf_{t\rightarrow 0} R(t)t^{-\frac12}= \infty$, and assuming higher regularity for the initial data, we obtain an improved lower bound near time zero. As an application, we prove that for any global solution~$u\in C([0,\infty); H^{\frac12}(\R^3))$ of the Navier-Stokes equations, there holds~$\displaystyle\liminf_{t\rightarrow \infty} R(t)t^{-\frac12}= \infty$.

math.AP↗

arXiv2020

MagGene: A genetic evolution program for magnetic structure prediction

We have developed a software MagGene to predict magnetic structures by using genetic algorithm. Starting from an atom structure, MagGene repeatedly generates new magnetic structures and calls first-principles calculation engine to get the most stable structure. This software is applicable to both collinear and noncollinear systems. It is particularly convenient for predicting the magnetic structures of atomic systems with strong spin-orbit couplings and/or strong spin frustrations.

cond-mat.mtrl-sci↗

arXiv2020

The Brownian Motion in an Ideal Quantum Qas

A Brownian particle in an ideal quantum gas is considered. The mean square displacement (MSD) is derived. The Bose-Einstein or Fermi-Dirac distribution, other than the Maxwell-Boltzmann distribution, provides a different stochastic force compared with the classical Brownian motion. The MSD, which depends on the thermal wavelength and the density of medium particles, reflects the quantum effect on the Brownian particle explicitly. The result shows that the MSD in an ideal Bose gas is shorter than that in a Fermi gas. The behavior of the quantum Brownian particle recovers the classical Brownian particle as the temperature raises. At low temperatures, the quantum effect becomes obvious. For example, there is a random motion of the Brownian particle due to the fermionic exchange interaction even the temperature is near the absolute zero.

cond-mat.stat-mech↗

arXiv2020

Graph Embedding on Biomedical Networks: Methods, Applications, and Evaluations

Graph embedding learning that aims to automatically learn low-dimensional node representations, has drawn increasing attention in recent years. To date, most recent graph embedding methods are evaluated on social and information networks and are not comprehensively studied on biomedical networks under systematic experiments and analyses. On the other hand, for a variety of biomedical network analysis tasks, traditional techniques such as matrix factorization (which can be seen as a type of graph embedding methods) have shown promising results, and hence there is a need to systematically evaluate the more recent graph embedding methods (e.g. random walk-based and neural network-based) in terms of their usability and potential to further the state-of-the-art. We select 11 representative graph embedding methods and conduct a systematic comparison on 3 important biomedical link prediction tasks: drug-disease association (DDA) prediction, drug-drug interaction (DDI) prediction, protein-protein interaction (PPI) prediction; and 2 node classification tasks: medical term semantic type classification, protein function prediction. Our experimental results demonstrate that the recent graph embedding methods achieve promising results and deserve more attention in the future biomedical graph analysis. Compared with three state-of-the-art methods for DDAs, DDIs and protein function predictions, the recent graph embedding methods achieve competitive performance without using any biological features and the learned embeddings can be treated as complementary representations for the biological features. By summarizing the experimental results, we provide general guidelines for properly selecting graph embedding methods and setting their hyper-parameters for different biomedical tasks.

cs.LG↗

arXiv2020

Global existence and decay of solutions to Prandtl system with small analytic data

In this paper, we prove the global existence and the large time decay estimate of solutions to Prandtl system with small initial data, which is analytical in the tangential variable. The key ingredient used in the proof is to derive sufficiently fast decay-in-time estimate of some weighted analytic energy estimate to a quantity, which consists of a linear combination of the tangential velocity with its primitive one, and which basically controls the evolution of the analytical radius to the solutions. Our result can be viewed as a global-in-time Cauchy-Kowalevsakya result for Prandtl system with small analytical data.

math.AP↗

arXiv2019

Semiclassical limit of Gross-Pitaevskii equation with Dirichlet boundary condition

In this paper, we justify the semiclassical limit of Gross-Pitaevskii equation with Dirichlet boundary condition on the 3-D upper space under the assumption that the leading order terms to both initial amplitude and initial phase function are sufficiently small in some high enough Sobolev norms. We remark that the main difficulty of the proof lies in the fact that the boundary layer appears in the leading order terms of the amplitude functions and the gradient of the phase functions to the WKB expansions of the solutions. In particular, we partially solved the open question proposed in \cite{CR2009, PNB2005} concerning the semiclassical limit of Gross-Pitaevskii equation with Dirichlet boundary condition.

math.AP↗

arXiv2019

Controllability of the Navier-Stokes equation in a rectangle with a little help of a distributed phantom force

We consider the 2D incompressible Navier-Stokes equation in a rectangle with the usual no-slip boundary condition prescribed on the upper and lower boundaries. We prove that for any positive time, for any finite energy initial data, there exist controls on the left and right boundaries and a distributed force, which can be chosen arbitrarily small in any Sobolev norm in space, such that the corresponding solution is at rest at the given final time. Our work improves earlier results where the distributed force is small only in a negative Sobolev space. It is a further step towards an answer to Jacques-Louis Lions' question about the small-time global exact boundary controllability of the Navier-Stokes equation with the no-slip boundary condition, for which no distributed force is allowed. Our analysis relies on the well-prepared dissipation method already used for Burgers and for Navier-Stokes in the case of the Navier slip-with-friction boundary condition. In order to handle the larger boundary layers associated with the no-slip boundary condition, we perform a preliminary regularization into analytic functions with arbitrarily large analytic radius and prove a long-time nonlinear Cauchy-Kovalevskaya estimate relying only on horizontal analyticity.

math.AP↗

arXiv2019

Extremal Problems in Royal Colorings of Graphs

An edge coloring $c$ of a graph $G$ is a royal $k$-edge coloring of $G$ if the edges of $G$ are assigned nonempty subsets of the set $\{1, 2, \ldots, k\}$ in such a way that the vertex coloring obtained by assigning the union of the colors of the incident edges of each vertex is a proper vertex coloring. If the vertex coloring is vertex-distinguishing, then $c$ is a strong royal $k$-edge coloring. The minimum positive integer $k$ for which $G$ has a strong royal $k$-edge coloring is the strong royal index of $G$. It has been conjectured that if $G$ is a connected graph of order $n\ge 4$ where $2^{k-1} \le n \le 2^k-1$ for a positive integer $k$, then the strong royal index of $G$ is either $k$ or $k+1$. We discuss this conjecture along with other information concerning strong royal colorings of graphs. A sufficient condition for such a graph to have a strong royal index $k+1$ is presented.

math.CO↗

arXiv2019

LabelECG: A Web-based Tool for Distributed Electrocardiogram Annotation

Electrocardiography plays an essential role in diagnosing and screening cardiovascular diseases in daily healthcare. Deep neural networks have shown the potentials to improve the accuracies of arrhythmia detection based on electrocardiograms (ECGs). However, more ECG records with ground truth are needed to promote the development and progression of deep learning techniques in automatic ECG analysis. Here we propose a web-based tool for ECG viewing and annotating, LabelECG. With the facilitation of unified data management, LabelECG is able to distribute large cohorts of ECGs to dozens of technicians and physicians, who can simultaneously make annotations through web-browsers on PCs, tablets and cell phones. Along with the doctors from four hospitals in China, we applied LabelECG to support the annotations of about 15,000 12-lead resting ECG records in three months. These annotated ECGs have successfully supported the First China ECG intelligent Competition. La-belECG will be freely accessible on the Internet to support similar researches, and will also be upgraded through future works.

cs.DB↗

arXiv2019

Van der Waals stacked multilayer in-plane graphene/hexagonal boron nitride heterostructure: its interfacial thermal transport properties

Combining both vertical and in-plane two-dimensional (2D) heterostructures opens up the possibility to create an unprecedented architecture using 2D atomic layer building blocks. The thermal transport properties of such mixed heterostructures, critical to various applications in nanoelectronics, however, have not been thoroughly explored. Herein, we construct two configurations of multilayer in-plane graphene/hexagonal boron nitride (Gr/h-BN) heterostructures (i.e. mixed heterostructures) via weak van der Waals (vdW) interactions and systematically investigate the dependence of their interfacial thermal conductance (ITC) on the number of layers using non-equilibrium molecular dynamics (NEMD) simulations. The computational results show that the ITC of two configurations of multilayer in-plane Gr/h-BN heterostructures (MIGHHs) decrease with increasing layer number n and both saturate at n = 3. And surprisingly, we find that the MIGHH is more advantageous to interfacial thermal transport than the monolayer in-plane Gr/h-BN heterostructure, which is in strong contrast to the commonly held notion that the multilayer structures of Gr and h-BN suppress the phonon transmission. The underlying physical mechanisms for these puzzling phenomena are probed through the analyses of heat flux, temperature jump, stress concentration factor, overlap of phonon vibrational spectra and phonon participation ratio. In particular, by changing the stacking angle of MIGHH, a higher ITC can be obtained due to the thermal rectification behavior. Furthermore, we find that the ITC in MIGHH can be well-regulated by controlling the coupling strength between layers. Our findings here are of significance for understanding the interfacial thermal transport behaviors of multilayer in-plane Gr/h-BN heterostructure, and are expected to attract extensive interest in exploring its new physics and applications.

cond-mat.mtrl-sci↗

arXiv2019

Age Based Task Scheduling and Computation Offloading in Mobile-Edge Computing Systems

To support emerging real-time monitoring and control applications, the timeliness of computation results is of critical importance to mobile-edge computing (MEC) systems. We propose a performance metric called age of task (AoT) based on the concept of age of information (AoI), to evaluate the temporal value of computation tasks. In this paper, we consider a system consisting of a single MEC server and one mobile device running several applications. We study an age minimization problem by jointly considering task scheduling, computation offloading and energy consumption. To solve the problem efficiently, we propose a light-weight task scheduling and computation offloading algorithm. Through performance evaluation, we show that our proposed age-based solution is competitive when compared with traditional strategies.

eess.SP↗

arXiv2019

Event-triggered Online Proactive Network Association to Mobile Edge Computing for IoT

Ultra-low latency communication for mobile machines emerges as a critical technology in Internet of Things (IoT). Proactive network association has been suggested to support ultra-low latency communication with the assistance of mobile edge computing. To resolve system dynamics and uncertainty, in this paper, an online proactive network association is proposed to minimize average task delay while considering time-average energy consumption constraints. Under distributed computing and networking environments, we formulate an event-triggered proactive network association model by semi-Markov task states and independent identically distributed (i.i.d.) random events. Then we facilitate the mobility-aware anticipatory network association to predictively consider handover effects caused by the mobility. Based on the Markov decision processes (MDP) and Lyapunov optimization, the two-stage online proactive network association (TOPNA) decision algorithm is proposed without the knowledge nor distribution of random events. Simulation results exhibit the effectiveness of the proposed algorithm.

cs.NI↗

arXiv2019

Global well-posedness of $3$-D anisotropic Navier-Stokes system with small unidirectional derivative

In \cite{LZ4}, the authors proved that as long as the one-directional derivative of the initial velocity is sufficiently small in some scaling invariant spaces, then the classical Navier-Stokes system has a global unique solution. The goal of this paper is to extend this type of result to the 3-D anisotropic Navier-Stokes system $(ANS)$ with only horizontal dissipation. More precisely, given initial data $u_0=(u_0^\h,u_0^3)\in \cB^{0,\f12},$ $(ANS)$ has a unique global solution provided that $|D_\h|^{-1}\pa_3u_0$ is sufficiently small in the scaling invariant space $\cB^{0,\f12}.$

math.AP↗