arXiv subjects

Meng Sun

Publications and source records attributed to Meng Sun.

At least 109 records · Page 6Linked to original sources

arXiv2019

Exciton-Polariton Topological Insulator with an Array of Magnetic Dots

Recently there have been several proposals on exciton-polariton topological insulators, most requiring strong external magnetic fields induced by bulky superconducting coils. We propose an alternate design for a polariton topological insulator, where excitons are in a proximity of an additional layer of a ferromagnetic material with a predefined magnetic moment located between the pillars of the cavity. Our design supports a variety of topological phases and transitions between Chern numbers $\pm 2$ and $\pm 1$ by varying either the magnetic moment of the ferromagnetic material or the spin-orbit coupling between different spin projections, thus enabling compact polariton devices harnessing switchable topological edge modes.

cond-mat.mes-hall↗

arXiv2019

GazeCorrection:Self-Guided Eye Manipulation in the wild using Self-Supervised Generative Adversarial Networks

Gaze correction aims to redirect the person's gaze into the camera by manipulating the eye region, and it can be considered as a specific image resynthesis problem. Gaze correction has a wide range of applications in real life, such as taking a picture with staring at the camera. In this paper, we propose a novel method that is based on the inpainting model to learn from the face image to fill in the missing eye regions with new contents representing corrected eye gaze. Moreover, our model does not require the training dataset labeled with the specific head pose and eye angle information, thus, the training data is easy to collect. To retain the identity information of the eye region in the original input, we propose a self-guided pretrained model to learn the angle-invariance feature. Experiments show our model achieves very compelling gaze-corrected results in the wild dataset which is collected from the website and will be introduced in details. Code is available at https://github.com/zhangqianhui/GazeCorrection.

cs.CV↗

arXiv2019

Decentralized Detection with Robust Information Privacy Protection

We consider a decentralized detection network whose aim is to infer a public hypothesis of interest. However, the raw sensor observations also allow the fusion center to infer private hypotheses that we wish to protect. We consider the case where there are an uncountable number of private hypotheses belonging to an uncertainty set, and develop local privacy mappings at every sensor so that the sanitized sensor information minimizes the Bayes error of detecting the public hypothesis at the fusion center, while achieving information privacy for all private hypotheses. We introduce the concept of a most favorable hypothesis (MFH) and show how to find a MFH in the set of private hypotheses. By protecting the information privacy of the MFH, information privacy for every other private hypothesis is also achieved. We provide an iterative algorithm to find the optimal local privacy mappings, and derive some theoretical properties of these privacy mappings. Simulation results demonstrate that our proposed approach allows the fusion center to infer the public hypothesis with low error while protecting information privacy of all the private hypotheses.

cs.IT↗

arXiv2019

On the Relationship Between Inference and Data Privacy in Decentralized IoT Networks

In a decentralized Internet of Things (IoT) network, a fusion center receives information from multiple sensors to infer a public hypothesis of interest. To prevent the fusion center from abusing the sensor information, each sensor sanitizes its local observation using a local privacy mapping, which is designed to achieve both inference privacy of a private hypothesis and data privacy of the sensor raw observations. Various inference and data privacy metrics have been proposed in the literature. We introduce the concepts of privacy implication and non-guarantee to study the relationships between these privacy metrics. We propose an optimization framework in which both local differential privacy (data privacy) and information privacy (inference privacy) metrics are incorporated. In the parametric case where sensor observations' distributions are known \emph{a priori}, we propose a two-stage local privacy mapping at each sensor, and show that such an architecture is able to achieve information privacy and local differential privacy to within the predefined budgets. For the nonparametric case where sensor distributions are unknown, we adopt an empirical optimization approach. Simulation and experiment results demonstrate that our proposed approaches allow the fusion center to accurately infer the public hypothesis while protecting both inference and data privacy.

cs.IT↗

arXiv2019

Unconventional Bloch-Grüneisen scattering in hybrid Bose-Fermi systems

We report on the novel mechanism of electron scattering in hybrid Bose-Fermi systems consisting of a two-dimensional electron gas in the vicinity of an exciton condensate: We show that a pair-of-bogolons--mediated scattering proves to be dominating over the conventional acoustic phonon channel and over the single-bogolon scattering, even if the screening is taken into account. We develop a microscopic theory of this effect, focusing on GaAs and MoS$_2$ materials, and find the principal temperature dependence of resistivity, distinct from the conventional phonon--mediated processes. Further, we scrutinize parameters and suggest a way to design composite samples with predefined electron mobilities and propose a mechanism of electron pairing for superconductivity.

cond-mat.mes-hall↗

arXiv2018

Bogolon-mediated electron scattering in graphene in hybrid Bose-Fermi systems

We report on an unconventional mechanism of electron scattering in graphene in hybrid Bose- Fermi systems. We study energy-dependent electron relaxation time, accounting for the processes of emission and absorption of a Bogoliubov excitation (a bogolon). Then using the Bloch-Gruneisen approach, we find the finite-temperature resistivity of graphene and show that its principal behavior is $\sim T^4$ in the limit of low temperatures and linear at high temperatures. We show that bogolon-mediated scattering can surpass the acoustic phonon-assisted relaxation. It can be controlled by the distance between the layers and the condensate density, giving us additional degrees of freedom and a useful tool to render electron mobility by the sample design and external pump.

cond-mat.mes-hall↗

arXiv2018

Orbital Decay in Binaries Containing Post-Main Sequence Stars

The orbital decay of binaries containing a primary sub-giant or red giant star and a stellar or substellar companion is investigated. The tide raised in the primary by the companion leads to an exchange of angular momentum between the orbit and the stellar spin, causing the orbit to contract and the primary to spin up. The rate of orbital decay is computed including both the equilibrium tide, damped by turbulent viscosity in the convective envelope, and the dynamical tide, assumed to be a traveling internal-gravity wave in the radiative core. For close binaries, the tidal forcing period is expected to be much shorter than the eddy turnover timescale in the convective envelope, and the prescription for "reduced" viscosity is an important consideration. The dynamical tide tends to dominate for the closest orbits, while the equilibrium tide dominates for more distant orbits, with the crossover point depending on the stellar mass. The spin up of the primary to synchronous rotation occurs for sufficiently massive secondaries, and this greatly slows the orbital decay until the Darwin instability occurs. A parameter survey is presented for orbital decay as a function of primary and secondary mass, as well as turbulent viscosity prescription. These results are summarized with analytic formulae and numerical results for the age-dependent critical separation, $a_{\rm crit}$, inside of which orbital decay is rapid, and few systems are expected to be observed. The calculations of $a_{\rm crit}$ are compared with APOGEE binaries, as well as solar mass exoplanet host stars.

astro-ph.SR↗

arXiv2018

Multi-Perspective Context Aggregation for Semi-supervised Cloze-style Reading Comprehension

Cloze-style reading comprehension has been a popular task for measuring the progress of natural language understanding in recent years. In this paper, we design a novel multi-perspective framework, which can be seen as the joint training of heterogeneous experts and aggregate context information from different perspectives. Each perspective is modeled by a simple aggregation module. The outputs of multiple aggregation modules are fed into a one-timestep pointer network to get the final answer. At the same time, to tackle the problem of insufficient labeled data, we propose an efficient sampling mechanism to automatically generate more training examples by matching the distribution of candidates between labeled and unlabeled data. We conduct our experiments on a recently released cloze-test dataset CLOTH (Xie et al., 2017), which consists of nearly 100k questions designed by professional teachers. Results show that our method achieves new state-of-the-art performance over previous strong baselines.

cs.CL↗

arXiv2018

Excitation of localized condensates in the flat band of exciton-polariton Lieb lattice

We propose a way to directly excite compact localized condensates in a nearly flat band of the exciton-polariton Lieb lattice by short Laguerre-Gaussian pulses and investigate the dynamics of these condensates in the presence of repulsive polariton-polariton interaction and distributed losses in the lattice. The evolution of a low-density compact polariton condensate shows fast Rabi oscillations between its excitonic and photonic components, with slow beatings of the Rabi oscillation amplitude. Both oscillations and beatings are suppressed at higher condensate densities due to polariton-polariton repulsion and distributed losses in the lattice. A background incoherent pumping can be used to increase the lifetime and stability of compact localized states

cond-mat.mes-hall↗

arXiv2018

Yuanfudao at SemEval-2018 Task 11: Three-way Attention and Relational Knowledge for Commonsense Machine Comprehension

This paper describes our system for SemEval-2018 Task 11: Machine Comprehension using Commonsense Knowledge. We use Three-way Attentive Networks (TriAN) to model interactions between the passage, question and answers. To incorporate commonsense knowledge, we augment the input with relation embedding from the graph of general knowledge ConceptNet (Speer et al., 2017). As a result, our system achieves state-of-the-art performance with 83.95% accuracy on the official test data. Code is publicly available at https://github.com/intfloat/commonsense-rc

cs.CL↗

arXiv2018

Multilayer Nonlinear Processing for Information Privacy in Sensor Networks

A sensor network wishes to transmit information to a fusion center to allow it to detect a public hypothesis, but at the same time prevent it from inferring a private hypothesis. We propose a multilayer nonlinear processing procedure at each sensor to distort the sensor's data before it is sent to the fusion center. In our proposed framework, sensors are grouped into clusters, and each sensor first applies a nonlinear fusion function on the information it receives from sensors in the same cluster and in a previous layer. A linear weighting matrix is then used to distort the information it sends to sensors in the next layer. We adopt a nonparametric approach and develop a modified mirror descent algorithm to optimize the weighting matrices so as to ensure that the regularized empirical risk of detecting the private hypothesis is above a given privacy threshold, while minimizing the regularized empirical risk of detecting the public hypothesis. Experiments on empirical datasets demonstrate that our approach is able to achieve a good trade-off between the error rates of the public and private hypothesis.

cs.CR↗

arXiv2018

Formation of Extremely Low-mass White Dwarfs Binaries

Motivated by the discovery of a handful of pulsating, extremely low mass white dwarfs (ELM WDs, mass $M \lesssim 0.17\, M_\odot$) which likely have WD companions, this paper discusses binary formation models for these systems. Formation of an ELM WD by unstable mass transfer (MT) or a common envelope (CE) event is unlikely. Stable Roche-lobe overflow with conservative MT produces only $M \gtrsim 0.2\, M_\odot$. This paper discusses the formation of ELM WD using angular momentum losses by magnetic braking. The initially more massive star becomes the companion WD through a CE event. The initially less massive star becomes the ELM WD as an evolved donor in a Cataclysmic Variable binary. Evolutionary models are constructed using the Modules for Experiments in Stellar Astrophysics (MESA), with ELM WD progenitors in the range $M_d=1.0-1.5\, M_\odot$ and WD companions in the range $M_{\rm a} \simeq 0.4-0.9\, M_\odot$. A prescription to reduce magnetic braking for thin surface convection zones is included. Upon the thinning of the evolved donor's envelope, the donor star shrinks out of contact and MT ceases, revealing the ELM WD. Systems with small masses have previously been suggested as possible AM CVN's. Systems with large masses, up to the limit $M \simeq 0.18\, M_\odot$ at which shell flashes occur on the WD cooling track, will tend to expand out to orbital periods $P_{\rm orb} \gtrsim 15\, {\rm hr}$. In between this range, ELM WD may become pulsators both as pre-WD and on the WD cooling track. Brickhill's criterion for convective mode driving is used to estimate the location of the blue edge of the g-mode instability strip.

astro-ph.SR↗

arXiv2017

Autoreactivity to malondialdehyde-modifications in rheumatoid arthritis is linked to disease activity and synovial pathogenesis

Oxidation-associated malondialdehyde (MDA) modification of proteins can generate immunogenic neo-epitopes that are recognized by autoantibodies. In health, IgM antibodies to MDA-adducts are part of the natural antibody pool, while elevated levels of IgG anti-MDA are associated with inflammatory conditions. Yet, in human autoimmune disease IgG anti-MDA responses have not been well characterized and their potential contribution to disease pathogenesis is not known. Here, we investigate MDA-modifications and anti-MDA-modified protein autoreactivity in rheumatoid arthritis (RA). While RA is primarily associated with autoreactivity to citrullinated antigens, we also observed increases in serum IgG anti-MDA in RA patients compared to controls. IgG anti-MDA levels significantly correlated with disease activity by DAS28-ESR and serum TNF-alpha, IL-6, and CRP. Mass spectrometry analysis of RA synovial tissue identified MDA-modified proteins and revealed shared peptides between MDA-modified and citrullinated actin and vimentin. Furthermore, anti-MDA autoreactivity among synovial B cells was discovered when investigating recombinant monoclonal antibodies (mAbs) cloned from single B cells. Several clones were highly specific for MDA-modification with no cross-reactivity to other antigen modifications. The mAbs recognized MDA-adducts in a variety of proteins. Interestingly, the most reactive clone, originated from an IgG1-bearing memory B cell, was encoded by germline variable genes, and showed similarity to previously reported natural IgM. Other anti-MDA clones display somatic hypermutations and lower reactivity. These anti-MDA antibodies had significant in vitro functional properties and induced enhanced osteoclastogenesis, while the natural antibody related high-reactivity clone did not. We postulate that these may represent distinctly different facets of anti-MDA autoreactive responses.

q-bio.BM↗

arXiv2017

Tidal Dissipation in WASP-12

WASP-12 is a hot Jupiter system with an orbital period of $P= 1.1\textrm{ day}$, making it one of the shortest-period giant planets known. Recent transit timing observations by Maciejewski et al. (2016) and Patra et al. (2017) find a decreasing period with $P/|\dot{P}| = 3.2\textrm{ Myr}$. This has been interpreted as evidence of either orbital decay due to tidal dissipation or a long term oscillation of the apparent period due to apsidal precession. Here we consider the possibility that it is orbital decay. We show that the parameters of the host star are consistent with either a $M_\ast \simeq 1.3 M_\odot$ main sequence star or a $M_\ast \simeq 1.2 M_\odot$ subgiant. We find that if the star is on the main sequence, the tidal dissipation is too inefficient to explain the observed $\dot{P}$. However, if it is a subgiant, the tidal dissipation is significantly enhanced due to nonlinear wave breaking of the dynamical tide near the star's center. The subgiant models have a tidal quality factor $Q_\ast'\simeq 2\times10^5$ and an orbital decay rate that agrees well with the observed $\dot{P}$. It would also explain why the planet survived for $\simeq 3\textrm{ Gyr}$ while the star was on the main sequence and yet is now inspiraling on a 3 Myr timescale. Although this suggests that we are witnessing the last $\sim 0.1\%$ of the planet's life, the probability of such a detection is a few percent given the observed sample of $\simeq 30$ hot Jupiters in $P<3\textrm{ day}$ orbits around $M_\ast>1.2 M_\odot$ hosts.

astro-ph.EP↗

arXiv2017

Towards Information Privacy for the Internet of Things

In an Internet of Things network, multiple sensors send information to a fusion center for it to infer a public hypothesis of interest. However, the same sensor information may be used by the fusion center to make inferences of a private nature that the sensors wish to protect. To model this, we adopt a decentralized hypothesis testing framework with binary public and private hypotheses. Each sensor makes a private observation and utilizes a local sensor decision rule or privacy mapping to summarize that observation independently of the other sensors. The local decision made by a sensor is then sent to the fusion center. Without assuming knowledge of the joint distribution of the sensor observations and hypotheses, we adopt a nonparametric learning approach to design local privacy mappings. We introduce the concept of an empirical normalized risk, which provides a theoretical guarantee for the network to achieve information privacy for the private hypothesis with high probability when the number of training samples is large. We develop iterative optimization algorithms to determine an appropriate privacy threshold and the best sensor privacy mappings, and show that they converge. Finally, we extend our approach to the case of a private multiple hypothesis. Numerical results on both synthetic and real data sets suggest that our proposed approach yields low error rates for inferring the public hypothesis, but high error rates for detecting the private hypothesis.

cs.IT↗

arXiv2016

Towards Concolic Testing for Hybrid Systems

Hybrid systems exhibit both continuous and discrete behavior. Analyzing hybrid systems is known to be hard. Inspired by the idea of concolic testing (of programs), we investigate whether we can combine random sampling and symbolic execution in order to effectively verify hybrid systems. We identify a sufficient condition under which such a combination is more effective than random sampling. Furthermore, we analyze different strategies of combining random sampling and symbolic execution and propose an algorithm which allows us to dynamically switch between them so as to reduce the overall cost. Our method has been implemented as a web-based checker named HyChecker. HyChecker has been evaluated with benchmark hybrid systems and a water treatment system in order to test its effectiveness.

cs.SE↗

arXiv2016

A Computational Model of YAP/TAZ Mechanosensing

In cell proliferation, stem cell differentiation, chemoresistance and tissue organization, the ubiquitous role of YAP/TAZ continues to impact our fundamental understanding in numerous physiological and disease systems. YAP/TAZ is an important signaling nexus integrating diverse mechanical and biochemical signals, such as ECM stiffness, adhesion ligand density, or cell-cell contacts, and thus strongly influences cell fate. Recent studies show that YAP/TAZ mechanical sensing is dependent on RhoA-regulated stress fibers. However, current understanding of YAP/TAZ still remains limited due to the unknown interaction between the canonical Hippo pathway and cell tension. To identify the roles of key signaling molecules in mechanical signal sensing and transduction, we present a novel computational model of the YAP/TAZ signaling pathway. This model converts ECM mechanical properties to biochemical signals via adhesion, and integrates intracellular signaling cascades associated with cytoskeleton dynamics. Adhesion molecules, such as FAK, are predicted to rescue YAP/TAZ activity in soft environments via the RhoA pathway. We found that changes of molecule concentrations result in different pattern of YAP/TAZ stiffness response. We also investigate the sensitivity of YAP/TAZ activity to ECM stiffness. In addition, the model shows that the unresolved synergistic effect of YAP/TAZ activity between the mechanosensing and the Hippo pathways can be explained by the interaction of LIMK and LATS. Overall, our model provides a novel platform for studying YAP/TAZ activity in the context of integrating different signaling pathways. This platform can be used to gain new fundamental insights into roles of key molecular and mechanical regulators on development, tissue engineering or tumor progression.

physics.bio-ph↗

arXiv2015

Thermodynamic fluctuation in black string flow

It has long been noticed that Laudau-Lifshitz theory can be used to study the fluctuation of a system that contains a black hole. Since the black string can be constructed by extending n-dimensional black hole into one extra dimension. We study the fluctuation of black string flow with a Schwarzschlid-like metric in D=n+1 dimensional spacetime and a charged solution in D=5 dimensional spacetime and get the second moments of the fluctuation of the mass flux and charge flux.

gr-qc↗