arXiv subjects

Yang Cao

Publications and source records attributed to Yang Cao.

At least 379 records · Page 21Linked to original sources

arXiv2018

Sketching for Sequential Change-Point Detection

We study sequential change-point detection procedures based on linear sketches of high-dimensional signal vectors using generalized likelihood ratio (GLR) statistics. The GLR statistics allow for an unknown post-change mean that represents an anomaly or novelty. We consider both fixed and time-varying projections, derive theoretical approximations to two fundamental performance metrics: the average run length (ARL) and the expected detection delay (EDD); these approximations are shown to be highly accurate by numerical simulations. We further characterize the relative performance measure of the sketching procedure compared to that without sketching and show that there can be little performance loss when the signal strength is sufficiently large, and enough number of sketches are used. Finally, we demonstrate the good performance of sketching procedures using simulation and real-data examples on solar flare detection and failure detection in power networks.

cs.LG↗

arXiv2018

Large tunable valley splitting in edge-free graphene quantum dots on boron nitride

Coherent manipulation of binary degrees of freedom is at the heart of modern quantum technologies. Graphene offers two binary degrees: the electron spin and the valley. Efficient spin control has been demonstrated in many solid state systems, while exploitation of the valley has only recently been started, yet without control on the single electron level. Here, we show that van-der Waals stacking of graphene onto hexagonal boron nitride offers a natural platform for valley control. We use a graphene quantum dot induced by the tip of a scanning tunneling microscope and demonstrate valley splitting that is tunable from -5 to +10 meV (including valley inversion) by sub-10-nm displacements of the quantum dot position. This boosts the range of controlled valley splitting by about one order of magnitude. The tunable inversion of spin and valley states should enable coherent superposition of these degrees of freedom as a first step towards graphene-based qubits.

cond-mat.mes-hall↗

arXiv2018

Comparing descent obstruction and Brauer-Manin obstruction for open varieties

We provide a relation between Brauer-Manin obstruction and descent obstruction for torsors over open varieties under a connected linear algebraic group or a group of multiplicative type is given. Such a relation is further refined for torsors under a torus. As an appliaction, we prove that the semi-simple part of a connected linear algebraic group will satisfy strong approximation with Brauer-Manin obstruction if G iteself satisfies strong approximation with Brauer-Manin obstruction.The equivalence between descent obstruction and etale Brauer-Manin obstruction for smooth projective varieties is extended to smooth quasi-projective varieties, which provides the perspective to study integral points.

math.NT↗

arXiv2018

Unusual suppression of the superconducting energy gap and critical temperature in atomically thin NbSe2

It is well known that superconductivity in thin films is generally suppressed with decreasing thickness. This suppression is normally governed by either disorder-induced localization of Cooper pairs, weakening of Coulomb screening, or generation and unbinding of vortex-antivortex pairs as described by the Berezinskii-Kosterlitz-Thouless (BKT) theory. Defying general expectations, few-layer NbSe2 - an archetypal example of ultrathin superconductors - has been found to remain superconducting down to monolayer thickness. Here we report measurements of both the superconducting energy gap and critical temperature in high-quality monocrystals of few-layer NbSe2, using planar-junction tunneling spectroscopy and lateral transport. We observe a fully developed gap that rapidly reduces for devices with the number of layers N < 5, as does their ctitical temperature. We show that the observed reduction cannot be explained by disorder, and the BKT mechanism is also excluded by measuring its transition temperature that for all N remains very close to Tc. We attribute the observed behavior to changes in the electronic band structure predicted for mono- and bi- layer NbSe2 combined with inevitable suppression of the Cooper pair density at the superconductor-vacuum interface. Our experimental results for N > 2 are in good agreement with the dependences of the gap and Tc expected in the latter case while the effect of band-structure reconstruction is evidenced by a stronger suppression of the gap and the disappearance of its anisotropy for N = 2. The spatial scale involved in the surface suppression of the density of states is only a few angstroms but cannot be ignored for atomically thin superconductors.

cond-mat.supr-con↗

arXiv2018

Change Detection via Affine and Quadratic Detectors

The goal of the paper is to develop a specific application of the convex optimization based hypothesis testing techniques developed in A. Juditsky, A. Nemirovski, "Hypothesis testing via affine detectors," Electronic Journal of Statistics 10:2204--2242, 2016. Namely, we consider the Change Detection problem as follows: given an evolving in time noisy observations of outputs of a discrete-time linear dynamical system, we intend to decide, in a sequential fashion, on the null hypothesis stating that the input to the system is a nuisance, vs. the alternative stating that the input is a "nontrivial signal," with both the nuisances and the nontrivial signals modeled as inputs belonging to finite unions of some given convex sets. Assuming the observation noises zero mean sub-Gaussian, we develop "computation-friendly" sequential decision rules and demonstrate that in our context these rules are provably near-optimal.

math.ST↗

arXiv2018

Multipath interference analysis of IR-UWB systems in indoor office LOS environment

Bit error rate (BER) performance of impulse radio Ultra-Wideband (UWB) systems in the presence of intrasymbol interference, inter-symbol interference, multiuser interference and addictive white Gaussian noise (AWGN) is presented in this paper. By analyzing the indoor office LOS channel model defined by IEEE 802.15.4a Task Group and deducing the variance for intra-symbol interference (IASI), inter-symbol interference (ISI) and multiuser interference (MUI), the system BER expression is obtained and verified by MATLAB simulations. Through comparing the simulation results with and without intra-symbol interference, the conclusion that intra-symbol interference cannot be neglected is drawn-moreover, such interference will significantly decrease performance of UWB based wireless sensor networks (WSN). Then, the BER performance of UWB systems in multiuser environment is also analyzed and analysis results show that multiuser interference will further worsen the transmission performance of UWB systems.

eess.SP↗

arXiv2018

Partial Template Based Receiver in Impulse Radio Ultra-Wideband Communication Systems

For high speed ultra-wideband (UWB) communication systems, the multipath interference exhibits a primary obstacle to improve transmission performance. In order to enhance the signal to interference plus noise ratio (SINR) in the receiver, a partial template receiver is proposed in this paper. Instead of using the conventional template, the model in this paper adopts a partial template to demodulate signals. To analyze the performance of the proposed receiver, bit error rate (BER) formulation of IR-UWB systems in the presence of multipath interference, multiuser interference and addictive white Gaussian noise (AWGN) is derived in IEEE 802.15.4a channel models. Simulation results show that, compared with the conventional correlation receiver, the proposed receiver can achieve a better BER performance for high Eb/N0, which falls in the conventional used Eb/N0 range.

eess.SP↗

arXiv2018

Performance Analysis of UWB Based Wireless Sensor Networks in Indoor Office LOS Environment

With the fast development of wireless sensor networks (WSN), more attentions are paid to high data rate transmission of WSN, and hence, in IEEE 802.15.4a standard, ultra-wideband (UWB) is introduced as one of the physical layer technique to support high transmission data rate and precisie locationing applications. In order to analyze the bit error rate (BER) performance of UWB based WSN, a system model considering intra-symbol interference (IASI), inter-symbol interference (ISI), multiuser interference (MUI) and addictive white Gaussian noise (AWGN) is proposed in this paper, and then verified using simulation results. Moreover, the pulse waveforms complying with the spectrum requirement of IEEE 802.15.4a standard are given, and based on such obtained pulses, the effect of transmission data rate and user number is also shown. Results show that with the increase of SNR, the intra-symbol interference will decrease the system performance significantly, and system performance can be improve by using pulse waveforms with little intra-symbol interference.

eess.SP↗

arXiv2018

How to Balance Privacy and Money through Pricing Mechanism in Personal Data Market

A personal data market is a platform including three participants: data owners (individuals), data buyers and market maker. Data owners who provide personal data are compensated according to their privacy loss. Data buyers can submit a query and pay for the result according to their desired accuracy. Market maker coordinates between data owner and buyer. This framework has been previously studied based on differential privacy. However, the previous study assumes data owners can accept any level of privacy loss and data buyers can conduct the transaction without regard to the financial budget. In this paper, we propose a practical personal data trading framework that is able to strike a balance between money and privacy. In order to gain insights on user preferences, we first conducted an online survey on human attitude to- ward privacy and interest in personal data trading. Second, we identify the 5 key principles of personal data market, which is important for designing a reasonable trading frame- work and pricing mechanism. Third, we propose a reason- able trading framework for personal data which provides an overview of how the data is traded. Fourth, we propose a balanced pricing mechanism which computes the query price for data buyers and compensation for data owners (whose data are utilized) as a function of their privacy loss. The main goal is to ensure a fair trading for both parties. Finally, we will conduct an experiment to evaluate the output of our proposed pricing mechanism in comparison with other previously proposed mechanism.

cs.CY↗

arXiv2017

Nearly second-order asymptotic optimality of sequential change-point detection with one-sample updates

Sequential change-point detection when the distribution parameters are unknown is a fundamental problem in statistics and machine learning. When the post-change parameters are unknown, we consider a set of detection procedures based on sequential likelihood ratios with non-anticipating estimators constructed using online convex optimization algorithms such as online mirror descent, which provides a more versatile approach to tackle complex situations where recursive maximum likelihood estimators cannot be found. When the underlying distributions belong to a exponential family and the estimators satisfy the logarithm regret property, we show that this approach is nearly second-order asymptotically optimal. This means that the upper bound for the false alarm rate of the algorithm (measured by the average-run-length) meets the lower bound asymptotically up to a log-log factor when the threshold tends to infinity. Our proof is achieved by making a connection between sequential change-point and online convex optimization and leveraging the logarithmic regret bound property of online mirror descent algorithm. Numerical and real data examples validate our theory.

math.ST↗

arXiv2017

Observing imperfection in atomic interfaces for van der Waals heterostructures

Vertically stacked van der Waals heterostructures are a lucrative platform for exploring the rich electronic and optoelectronic phenomena in two-dimensional materials. Their performance will be strongly affected by impurities and defects at the interfaces. Here we present the first systematic study of interfaces in van der Waals heterostructure using cross sectional scanning transmission electron microscope (STEM) imaging. By measuring interlayer separations and comparing these to density functional theory (DFT) calculations we find that pristine interfaces exist between hBN and MoS2 or WS2 for stacks prepared by mechanical exfoliation in air. However, for two technologically important transition metal dichalcogenide (TMDC) systems, MoSe2 and WSe2, our measurement of interlayer separations provide the first evidence for impurity species being trapped at buried interfaces with hBN: interfaces which are flat at the nanometer length scale. While decreasing the thickness of encapsulated WSe2 from bulk to monolayer we see a systematic increase in the interlayer separation. We attribute these differences to the thinnest TMDC flakes being flexible and hence able to deform mechanically around a sparse population of protruding interfacial impurities. We show that the air sensitive two dimensional (2D) crystal NbSe2 can be fabricated into heterostructures with pristine interfaces by processing in an inert-gas environment. Finally we find that adopting glove-box transfer significantly improves the quality of interfaces for WSe2 compared to processing in air.

cond-mat.mes-hall↗

arXiv2017

Robust Sequential Change-Point Detection by Convex Optimization

We address the computational challenge of finding the robust sequential change-point detection procedures when the pre- and post-change distributions are not completely specified. Earlier works [veeravalli 1994] and [Unnikrishnan 2011] establish the general conditions for robust procedures which include finding a pair of least favorable distributions (LFDs). However, in the multi-dimensional setting, it is hard to find such LFDs computationally. We present a method based on convex optimization that addresses this issue when the distributions are Gaussian with unknown parameters from pre-specified uncertainty sets. We also establish theoretical properties of our robust procedures, and numerical examples demonstrate their good performance.

stat.ME↗

arXiv2017

Quantifying Differential Privacy under Temporal Correlations

Differential Privacy (DP) has received increased attention as a rigorous privacy framework. Existing studies employ traditional DP mechanisms (e.g., the Laplace mechanism) as primitives, which assume that the data are independent, or that adversaries do not have knowledge of the data correlations. However, continuously generated data in the real world tend to be temporally correlated, and such correlations can be acquired by adversaries. In this paper, we investigate the potential privacy loss of a traditional DP mechanism under temporal correlations in the context of continuous data release. First, we model the temporal correlations using Markov model and analyze the privacy leakage of a DP mechanism when adversaries have knowledge of such temporal correlations. Our analysis reveals that the privacy leakage of a DP mechanism may accumulate and increase over time. We call it temporal privacy leakage. Second, to measure such privacy leakage, we design an efficient algorithm for calculating it in polynomial time. Although the temporal privacy leakage may increase over time, we also show that its supremum may exist in some cases. Third, to bound the privacy loss, we propose mechanisms that convert any existing DP mechanism into one against temporal privacy leakage. Experiments with synthetic data confirm that our approach is efficient and effective.

cs.DB↗

arXiv2017

Tuning the pseudospin polarization of graphene by a pseudo-magnetic field

One of the intriguing characteristics of honeycomb lattices is the appearance of a pseudo-magnetic field as a result of mechanical deformation. In the case of graphene, the Landau quantization resulting from this pseudo-magnetic field has been measured using scanning tunneling microscopy. Here we show that a signature of the pseudo-magnetic field is a local sublattice symmetry breaking observable as a redistribution of the local density of states. This can be interpreted as a polarization of graphene's pseudospin due to a strain induced pseudo-magnetic field, in analogy to the alignment of a real spin in a magnetic field. We reveal this sublattice symmetry breaking by tunably straining graphene using the tip of a scanning tunneling microscope. The tip locally lifts the graphene membrane from a SiO$_2$ support, as visible by an increased slope of the $I(z)$ curves. The amount of lifting is consistent with molecular dynamics calculations, which reveal a deformed graphene area under the tip in the shape of a Gaussian. The pseudo-magnetic field induced by the deformation becomes visible as a sublattice symmetry breaking which scales with the lifting height of the strained deformation and therefore with the pseudo-magnetic field strength. Its magnitude is quantitatively reproduced by analytic and tight-binding models, revealing fields of 1000 T. These results might be the starting point for an effective THz valley filter, as a basic element of valleytronics.

cond-mat.mes-hall↗

arXiv2016

Approximation forte pour les variétés avec une action d'un groupe linéaire

Let $G$ be a connected linear algebraic group over a number field. Let $U \hookrightarrow X$ be a $G$-equivariant open embedding of a $G$-homogeneous space with connected stabilizers into a smooth $G$-variety. We prove that $X$ satisfies strong approximation with Brauer-Manin condition off a set $S$ of places of $k$ under either of the following hypotheses : (i) $S$ is the set of archimedean places; (ii) $S$ is a nonempty finite set and $\bar{k}^{\times}= \bar{k}[X]^{\times}$. The proof builds upon the case $X=U$, which has been the object of several works.

math.AG↗

arXiv2016

Random Knockout Tournaments

We consider a random knockout tournament among players $1, \ldots, n$, in which each match involves two players. The match format is specified by the number of matches played in each round, where the constitution of the matches in a round is random. Supposing that there are numbers $v_1, \ldots, v_n$ such that a match between $i$ and $j$ will be won by $i$ with probability $\frac{v_i}{v_i+v_j}$, we obtain a lower bound on the tournament win probability for the best player, as well as upper and lower bounds for all the players. We also obtain additional bounds by considering the best and worst formats for player $1$ in the special case $v_1 > v_2 = v_3 = \cdots = v_n.$

math.PR↗

arXiv2016

Stacking transition in bilayer graphene caused by thermally activated rotation

Crystallographic alignment between two-dimensional crystals in van der Waals heterostructures brought a number of profound physical phenomena, including observation of Hofstadter butterfly and topological currents, and promising novel applications, such as resonant tunnelling transistors. Here, by probing the electronic density of states in graphene using graphene-hexagonal boron nitride tunnelling transistors, we demonstrate a structural transition of bilayer graphene from incommensurate twisted stacking state into a commensurate AB stacking due to a macroscopic graphene self-rotation. This structural transition is accompanied by a topological transition in the reciprocal space and by pseudospin texturing. The stacking transition is driven by van der Waals interaction energy of the two graphene layers and is thermally activated by unpinning the microscopic chemical adsorbents which are then removed by the self-cleaning of graphene.

cond-mat.mes-hall↗

arXiv2016

Cohomologie non ramifiée de degré 3 : variétés cellulaires et surfaces de del Pezzo de degré au moins 5

We consider geometrically cellular varieties $X$ over an arbitrary field of characteristic zero. We study the quotient of the third unramified cohomology group $H^3_{nr}(X,\mathbb{Q}/\mathbb{Z}(2))$ by its constant part. For $X$ a smooth compactification of a universal torsor over a geometrically rational surface, we show that this quotient if finite. For $X$ a del Pezzo surface of degree $\geq 5$, we show that this quotient is zero, unless $X$ is a del Pezzo surface of degree 8 of a special type.

math.AG↗