arXiv subjects

Rui Li

Publications and source records attributed to Rui Li.

At least 505 records · Page 28Linked to original sources

arXiv2019

Slowing-down reduction and Possible Reversal Trend of Tropospheric NO2 over China during 2016 to 2019

Atmospheric nitrogen dioxide (NO2) over China at national level has been kept reducing since 2011 as seen from both satellite observations, ground-based measurements and bottom-up emission inventory (Liu et al., 2016; Irie et al., 2016; Krotkov et al., 2016; Foy et al., 2016; Liu et al 2017). These studies used data before 2015-2016. After 2016, however, a significant slowing-down of the reduction trend and/or even a reversal trend were found in numerous provinces, particularly in those with heavy NO2 level, based on satellite observations. Error analysis on satellite data excluded cloud contamination, instrument anomalies from the main reasons of this change. Ground-based measurements show strong positive correlations with satellite observations and similar patterns of year-to-year changes of NO2 in 2018 and 2019 winter time. The temporal variations of Satellite NO2 over China are believed largely determined by surface emission from power plant and transportation. The reason for the recent change from emission perspective cannot be traced down since the national emission inventory was not updated since 2015. We therefore call on immediate attentions from both scientist community and policy makers to this phenomenon. Further efforts should be made to understand the reasons causing this change and to make associated air pollution controlling actions.

physics.ao-ph↗

arXiv2019

Neural Embedding for Physical Manipulations

In common real-world robotic operations, action and state spaces can be vast and sometimes unknown, and observations are often relatively sparse. How do we learn the full topology of action and state spaces when given only few and sparse observations? Inspired by the properties of grid cells in mammalian brains, we build a generative model that enforces a normalized pairwise distance constraint between the latent space and output space to achieve data-efficient discovery of output spaces. This method achieves substantially better results than prior generative models, such as Generative Adversarial Networks (GANs) and Variational Auto-Encoders (VAEs). Prior models have the common issue of mode collapse and thus fail to explore the full topology of output space. We demonstrate the effectiveness of our model on various datasets both qualitatively and quantitatively.

cs.LG↗

arXiv2019

Use of OWL and Semantic Web Technologies at Pinterest

Pinterest is a popular Web application that has over 250 million active users. It is a visual discovery engine for finding ideas for recipes, fashion, weddings, home decoration, and much more. In the last year, the company adopted Semantic Web technologies to create a knowledge graph that aims to represent the vast amount of content and users on Pinterest, to help both content recommendation and ads targeting. In this paper, we present the engineering of an OWL ontology---the Pinterest Taxonomy---that forms the core of Pinterest's knowledge graph, the Pinterest Taste Graph. We describe modeling choices and enhancements to WebProtégé that we used for the creation of the ontology. In two months, eight Pinterest engineers, without prior experience of OWL and WebProtégé, revamped an existing taxonomy of noisy terms into an OWL ontology. We share our experience and present the key aspects of our work that we believe will be useful for others working in this area.

cs.CL↗

arXiv2019

Spin manipulation and spin dephasing in quantum dot integrated with a slanting magnetic field

A slanting magnetic field is usually used to realize a slight hybridization between the spin and orbital degrees of freedom in a semiconductor quantum dot, such that the spin is manipulable by an external oscillating electric field. Here we show that, the longitudinal slanting field mediates a longitudinal driving term in the electric-dipole spin resonance, such that the spin population inversion exhibits a modulated Rabi oscillation. Fortunately, we can reduce this modulation by increasing the static magnetic field. The longitudinal slanting field also mediates a spin-1/f-charge noise interaction, which causes the pure dephasing of the spin qubit. Choosing proper spectrum function strength, we find the spin dephasing time is about $T^{*}_{2}=20$ $μ$s and the spin echo time is about $T^{\rm echo}_{2}=100$ $μ$s in a Si quantum dot. We also propose several strategies to alleviate the spin dephasing, such as lowering the experimental temperature, reducing the quantum dot size, engineering the slanting field, or using the dynamical decoupling scheme.

cond-mat.mes-hall↗

arXiv2019

Max-Min Fair Resource Allocation in Millimetre-Wave Backhauls

5G mobile networks are expected to provide pervasive high speed wireless connectivity, to support increasingly resource intensive user applications. Network hyper-densification therefore becomes necessary, though connecting to the Internet tens of thousands of base stations is non-trivial, especially in urban scenarios where optical fibre is difficult and costly to deploy. The millimetre wave (mm-wave) spectrum is a promising candidate for inexpensive multi-Gbps wireless backhauling, but exploiting this band for effective multi-hop data communications is challenging. In particular, resource allocation and scheduling of very narrow transmission/ reception beams requires to overcome terminal deafness and link blockage problems, while managing fairness issues that arise when flows encounter dissimilar competition and traverse different numbers of links with heterogeneous quality. In this paper, we propose WiHaul, an airtime allocation and scheduling mechanism that overcomes these challenges specific to multi-hop mm-wave networks, guarantees max-min fairness among traffic flows, and ensures the overall available backhaul resources are fully utilised. We evaluate the proposed WiHaul scheme over a broad range of practical network conditions, and demonstrate up to 5 times individual throughput gains and a fivefold improvement in terms of measurable fairness, over recent mm-wave scheduling solutions.

cs.NI↗

arXiv2019

Constraining the Rastall parameters in static spacetimes with galaxy-scale strong gravitational lensing

Recently, Rastall gravity is undergoing a significant surge in popularity. We obtain a power-law total mass-density profile for the inner region (within several effective radius) of early-type galaxies (ETGs) from the space-time structures which are described by the static spherically-symmetric solutions of Rastall gravity under the assumption of perfect fluid matter. We find that in the inner region of ETGs, the Rastall dimensionless parameter $β=κλ$ determines the mass distribution. We then use 118 galaxy-galaxy strong gravitational lensing systems to constrain the Rastall dimensionless parameter $β$. We find that the mean value of $β$ for total 118 ETGs is $β=0.163\pm0.001$(68\% CL) with a minor intrinsic scatter of $δ=0.020\pm 0.001$. Our work observationally illustrates the physical meaning of the Rastall dimensionless parameter in galaxy scale. From the Newtonian approximation of Rastall gravity, we also find that an absolute isothermal mass distribution for ETGs is not allowed in the framework of Rastall gravity.

astro-ph.GA↗

arXiv2019

Deep Distribution Regression

Due to their flexibility and predictive performance, machine-learning based regression methods have become an important tool for predictive modeling and forecasting. However, most methods focus on estimating the conditional mean or specific quantiles of the target quantity and do not provide the full conditional distribution, which contains uncertainty information that might be crucial for decision making. In this article, we provide a general solution by transforming a conditional distribution estimation problem into a constrained multi-class classification problem, in which tools such as deep neural networks. We propose a novel joint binary cross-entropy loss function to accomplish this goal. We demonstrate its performance in various simulation studies comparing to state-of-the-art competing methods. Additionally, our method shows improved accuracy in a probabilistic solar energy forecasting problem.

stat.ML↗

arXiv2019

Interfacial and Electronic Properties of Heterostructures of MXene and Graphene

MXene-based heterostructures have received considerable interest owing to their unique properties. Herein, we examine various heterostructures of a prototypical MXene and graphene using density functional theory. We find that the adhesion energy, charge transfer, and band structure of these heterostructures are sensitive not only to the surface functional group, but also to the stacking order. Difference in work function dictates the direction and amount of electron transfer across the interface, which causes a shift in the Dirac point of the graphene bands in the heterostructures of monolayer graphene and monolayer MXene. In the heterostructures of bilayer graphene and monolayer MXene, the interface breaks the symmetry of the bilayer graphene; in the case of the AB-stacking bilayer, the electron transfer leads to an interfacial electric field that opens up a gap in the graphene bands at the K point. This internal polarization strengthens both the interfacial adhesions and the cohesion between the two graphene layers. The MXene-graphene-MXene and graphene-MXene-graphene sandwich structures behave as two mirror-symmetric MXene-graphene interfaces. Our first principles studies provide a comprehensive understanding for the interaction between a typical MXene and graphene.

cond-mat.mes-hall↗

arXiv2018

Some monotone properties for solutions to a reaction-diffusion model

Motivated by the recent investigation of a predator-prey model in heterogeneous environments \cite{LouYuan-WangBiao}, we show that the maximum of the unique positive solution of the scalar equation \begin{equation}\label{eq:01}\begin{cases} μΔθ+(m(x)-θ)θ=0 \hspace{0.5em} &\text{in}\hspace{0.5em}Ω,\\ \frac{\partial θ}{\partial n}=0 \hspace{0.5em} &\text{on}\hspace{0.5em}\partialΩ\end{cases}\end{equation} is a strictly monotone decreasing function of the diffusion rate $μ$ for several classes of function $m$, which substantially improves a result in \cite{LouYuan-WangBiao}. However, the minimum of the positive solution of \eqref{eq:01} is monotone increasing under proper assumptions on the resource function, it is not always monotone increasing in the diffusion rate \cite{HeXiaoqing-NiWeiMing2016}.

math.AP↗

arXiv2018

Towards Machine Learning Prediction of Deep Brain Stimulation (DBS) Intra-operative Efficacy Maps

Deep brain stimulation (DBS) has the potential to improve the quality of life of people with a variety of neurological diseases. A key challenge in DBS is in the placement of a stimulation electrode in the anatomical location that maximizes efficacy and minimizes side effects. Pre-operative localization of the optimal stimulation zone can reduce surgical times and morbidity. Current methods of producing efficacy probability maps follow an anatomical guidance on magnetic resonance imaging (MRI) to identify the areas with the highest efficacy in a population. In this work, we propose to revisit this problem as a classification problem, where each voxel in the MRI is a sample informed by the surrounding anatomy. We use a patch-based convolutional neural network to classify a stimulation coordinate as having a positive reduction in symptoms during surgery. We use a cohort of 187 patients with a total of 2,869 stimulation coordinates, upon which 3D patches were extracted and associated with an efficacy score. We compare our results with a registration-based method of surgical planning. We show an improvement in the classification of intraoperative stimulation coordinates as a positive response in reduction of symptoms with AUC of 0.670 compared to a baseline registration-based approach, which achieves an AUC of 0.627 (p < 0.01). Although additional validation is needed, the proposed classification framework and deep learning method appear well-suited for improving pre-surgical planning and personalize treatment strategies.

cs.CV↗

arXiv2018

A scenario on the power-law total mass-density profile of the inner regions of early-type-galaxies

We naturally obtain a power-law total mass-density profile for the inner regions (within several effective radius) of early-type galaxies(ETGs) from space-time structures, which are described by the static spherically-symmetric solutions of Einstein equations with the perfect fluid matters. We point out that, in the inner region of an early type galaxy, the average value of the equation of state $ω$ determines the power-law index $γ$ of the total mass-density profile given by \cite{2006ApJ...649..599K}. We then use 119 galaxy-galaxy strong gravitational lensing systems from SLACS, BELLS, BELLS GALLERY and SL2S, to constrain the equation of state $ω$ for the inner regions of ETGs. We find that the average value of $ω$ for totally 119 ETGs is $ω=-0.315\pm0.073$ (68\% CL), which is very close to $-1/3$. If $ω=-1/3$ is the best value to describe the equation of state of the inner regions of the dark matter in galaxies, the deviation from $-1/3$ could represent the mixture of baryonic matter and dark matter in galaxies.

astro-ph.GA↗

arXiv2018

Using deep Residual Networks to search for galaxy-Lyα emitter lens candidates based on spectroscopic-selection

More than one hundred galaxy-scale strong gravitational lens systems have been found by searching for the emission lines coming from galaxies with redshifts higher than the lens galaxies. Based on this spectroscopic-selection method, we introduce the deep Residual Networks (ResNet, a kind of deep Convolutional Neural Networks) to search for the galaxy-Ly$α$ emitter (LAE) lens candidates by recognizing the Ly$α$ emission lines coming from high redshift galaxies ($2 < z < 3$) in the spectra of early-type galaxies (ETGs) at middle redshift ($z\sim 0.5$). The spectra of the ETGs come from the Data Release 12 (DR12) of the Baryon Oscillation Spectroscopic Survey (BOSS) of the Sloan Digital Sky Survey \uppercase\expandafter{\romannumeral3} (SDSS-\uppercase\expandafter{\romannumeral3}). In this paper, we first build a 28 layers ResNet model, and then artificially synthesize 150,000 training spectra, including 140,000 spectra without Ly$α$ lines and 10,000 ones with Ly$α$ lines, to train the networks. After 20 training epochs, we obtain a near-perfect test accuracy at 0.9954. The corresponding loss is 0.0028 and the completeness is 93.6\%. We finally apply our ResNet model to our predictive data with 174 known lens candidates. We obtain 1232 hits including 161 of the 174 known candidates (92.5\% discovery rate). Apart from the hits found in other works, our ResNet model also find 536 new hits. We then perform several subsequent selections on these 536 hits and present 5 most believable lens candidates.

astro-ph.GA↗

arXiv2018

DELMU: A Deep Learning Approach to Maximising the Utility of Virtualised Millimetre-Wave Backhauls

Advances in network programmability enable operators to 'slice' the physical infrastructure into independent logical networks. By this approach, each network slice aims to accommodate the demands of increasingly diverse services. However, precise allocation of resources to slices across future 5G millimetre-wave backhaul networks, to optimise the total network utility, is challenging. This is because the performance of different services often depends on conflicting requirements, including bandwidth, sensitivity to delay, or the monetary value of the traffic incurred. In this paper, we put forward a general rate utility framework for slicing mm-wave backhaul links, encompassing all known types of service utilities, i.e. logarithmic, sigmoid, polynomial, and linear. We then introduce DELMU, a deep learning solution that tackles the complexity of optimising non-convex objective functions built upon arbitrary combinations of such utilities. Specifically, by employing a stack of convolutional blocks, DELMU can learn correlations between traffic demands and achievable optimal rate assignments. We further regulate the inferences made by the neural network through a simple 'sanity check' routine, which guarantees both flow rate admissibility within the network's capacity region and minimum service levels. The proposed method can be trained within minutes, following which it computes rate allocations that match those obtained with state-of-the-art global optimisation algorithms, yet orders of magnitude faster. This confirms the applicability of DELMU to highly dynamic traffic regimes and we demonstrate up to 62% network utility gains over a baseline greedy approach.

cs.NI↗

arXiv2018

A spin dephasing mechanism mediated by the interplay between the spin-orbit coupling and the asymmetrical confining potential in semiconductor quantum dot

Understanding the spin dephasing mechanism is of fundamental importance in all potential applications of the spin qubit. Here we demonstrate a spin dephasing mechanism in semiconductor quantum dot due to the $1/f$ charge noise. The spin-charge interaction is mediated by the interplay between the spin-orbit coupling and the asymmetrical quantum dot confining potential. The dephasing rate is proportional to both the strength of the spin-orbit coupling and the degree of the asymmetry of the confining potential. For parameters typical of the InSb, InAs, and GaAs quantum dots with a moderate well-height $V_{0}=10$ meV, we find the spin dephasing times are ${\rm T}^{*}_{2}=7$ $μ$s, $275$ $μ$s, and $55$ ms, respectively. In particular, the spin dephasing can be enhanced by lowering the well-height. When the well-height is as small as $V_{0}=5$ meV, the spin depahsing times in the InSb, InAs, and GaAs quantum dots are decreased to ${\rm T}^{*}_{2}=0.38$ $μ$s, $18$ $μ$s, and $9$ ms, respectively.

cond-mat.mes-hall↗

arXiv2018

Toward A Deep Understanding of What Makes a Scientific Visualization Memorable

We report results from a preliminary study exploring the memorability of spatial scientific visualizations, the goal of which is to understand the visual features that contribute to memorability. The evaluation metrics include three objective measures (entropy, feature congestion, the number of edges), four subjective ratings (clutter, the number of distinct colors, familiarity, and realism), and two sentiment ratings (interestingness and happiness). We curate 1142 scientific visualization (SciVis) images from the original 2231 images in published IEEE SciVis papers from 2008 to 2017 and compute memorability scores of 228 SciVis images from data collected on Amazon Mechanical Turk (MTurk). Results showed that the memorability of SciVis images is mostly correlated with clutter and the number of distinct colors. We further investigate the differences between scientific visualization and infographics as a means to understand memorability differences by data attributes.

cs.GR↗

arXiv2018

A parabolic-hyperbolic system modeling the growth of a tumor

In this paper, we consider a model with tumor microenvironment involving nutrient density, extracellular matrix and matrix-degrading enzymes, which satisfy a coupled system of PDEs with a free boundary. For this coupled parabolic-hyperbolic free boundary problem, we prove that there is a unique radially symmetric solution globally in time. The stationary problem involves an ODE system which is transformed into a singular integro-differential equation. We establish a well-posed theorem for such general types of equations by the shooting method; the theorem is then applied to our problem for the existence of a stationary solution. In addition, for this highly nonlinear problem, we also prove the uniqueness of the stationary solution, which is a nontrivial result. In addition, numerical simulations indicate that the stationary solution is likely locally asymptotically stable for reasonable range of parameters.

math.AP↗

arXiv2018

Spin-relaxation anisotropy in a nanowire quantum dot with strong spin-orbit coupling

We study the impacts of the magnetic field direction on the spin-manipulation and the spin-relaxation in a one-dimensional quantum dot with strong spin-orbit coupling. The energy spectrum and the corresponding eigenfunctions in the quantum dot are obtained exactly. We find that no matter how large the spin-orbit coupling is, the electric-dipole spin transition rate as a function of the magnetic field direction always has a $π$ periodicity. However, the phonon-induced spin relaxation rate as a function of the magnetic field direction has a $π$ periodicity only in the weak spin-orbit coupling regime, and the periodicity is prolonged to $2π$ in the strong spin-orbit coupling regime.

cond-mat.mes-hall↗

arXiv2018

Strong-lensing Measurement of the Mass-density Profile out to 3 Effective Radii for $z \sim 0.5$ Early-type Galaxies

We measure the total mass-density profiles out to three effective radii for a sample of 63 $z \sim 0.5$, massive early-type galaxies (ETGs) acting as strong gravitational lenses through a joint analysis of lensing and stellar dynamics. The compilation is selected from three galaxy-scale strong-lens samples including the Baryon Oscillation Spectroscopic Survey (BOSS) Emission-Line Lens Survey (BELLS), BELLS for GALaxy-Ly$α$ EmitteR sYstems Survey, and Strong Lensing Legacy Survey (SL2S). Utilizing the wide source-redshift coverage (0.8--3.5) provided by these three samples, we build a statistically significant ensemble of massive ETGs for which robust mass measurements can be achieved within a broad range of Einstein radii up to three effective radii. Characterizing the three-dimensional total mass-density distribution by a power-law profile as $ρ\propto r^{-γ}$, we find that the average logarithmic density slope for the entire sample is $\langleγ\rangle=2.000_{-0.032}^{+0.033}$ ($68\%$CL) with an intrinsic scatter $δ=0.180_{-0.028}^{+0.032}$. Further parameterizing $\langleγ\rangle$ as a function of redshift $z$ and ratio of Einstein radius to effective radius $R_{ein}/R_{eff}$, we find the average density distributions of these massive ETGs become steeper at larger radii and later cosmic times with magnitudes $\mathrm{d} \langleγ\rangle / \mathrm{d}z = -0.309_{-0.160}^{+0.166}$ and $\mathrm{d} \langleγ\rangle / \mathrm{d} \log_{10} \frac{R_{ein}}{R_{eff}} = 0.194_{-0.083}^{+0.092}$.

astro-ph.GA↗