arXiv subjects

Di Wang

Publications and source records attributed to Di Wang.

At least 613 records · Page 34Linked to original sources

arXiv2018

Differentially Private Empirical Risk Minimization Revisited: Faster and More General

In this paper we study the differentially private Empirical Risk Minimization (ERM) problem in different settings. For smooth (strongly) convex loss function with or without (non)-smooth regularization, we give algorithms that achieve either optimal or near optimal utility bounds with less gradient complexity compared with previous work. For ERM with smooth convex loss function in high-dimensional ($p\gg n$) setting, we give an algorithm which achieves the upper bound with less gradient complexity than previous ones. At last, we generalize the expected excess empirical risk from convex loss functions to non-convex ones satisfying the Polyak-Lojasiewicz condition and give a tighter upper bound on the utility than the one in \cite{ijcai2017-548}.

cs.LG↗

arXiv2018

Large Scale Constrained Linear Regression Revisited: Faster Algorithms via Preconditioning

In this paper, we revisit the large-scale constrained linear regression problem and propose faster methods based on some recent developments in sketching and optimization. Our algorithms combine (accelerated) mini-batch SGD with a new method called two-step preconditioning to achieve an approximate solution with a time complexity lower than that of the state-of-the-art techniques for the low precision case. Our idea can also be extended to the high precision case, which gives an alternative implementation to the Iterative Hessian Sketch (IHS) method with significantly improved time complexity. Experiments on benchmark and synthetic datasets suggest that our methods indeed outperform existing ones considerably in both the low and high precision cases.

cs.LG↗

arXiv2018

Exploiting Investors Social Network for Stock Prediction in China's Market

Recent works have shown that social media platforms are able to influence the trends of stock price movements. However, existing works have majorly focused on the U.S. stock market and lacked attention to certain emerging countries such as China, where retail investors dominate the market. In this regard, as retail investors are prone to be influenced by news or other social media, psychological and behavioral features extracted from social media platforms are thought to well predict stock price movements in the China's market. Recent advances in the investor social network in China enables the extraction of such features from web-scale data. In this paper, on the basis of tweets from Xueqiu, a popular Chinese Twitter-like social platform specialized for investors, we analyze features with regard to collective sentiment and perception on stock relatedness and predict stock price movements by employing nonlinear models. The features of interest prove to be effective in our experiments.

cs.CE↗

arXiv2017

Spin Quenching Assisted by a Strongly Anisotropic Compression Behavior in MnP

We studied the crystal structure and spin state of MnP under high pressure with synchrotron X-ray diffraction and X-ray emission spectroscopy. MnP has an exceedingly strong anisotropy in compressibility, with the primary compressible direction along the b axis of the Pnma structure. X-ray emission spectroscopy reveals a pressure-driven quenching of the spin state in MnP. Firstprinciples calculations suggest that the strongly anisotropic compression behavior significantly enhances the dispersion of the Mn d-orbitals and the splitting of the d orbital levels compared to the hypothetical isotropic compression behavior. Thus, we propose spin quenching results mainly from the significant enhancement of the itinerancy of d electrons and partly from spin rearrangement occurring in the split d-orbital levels near the Fermi level. This explains the fast suppression of magnetic ordering in MnP under high pressure. The spin quenching lags behind the occurrence of superconductivity at ~8 GPa implying that spin fluctuations govern the electron pairing for superconductivity.

cond-mat.supr-con↗

arXiv2017

$CP$ asymmetries in charm decays into neutral kaons

We find a new $CP$-violation effect in charm decays into neutral kaons, which results from the interference between two tree (Cabibbo-favored and doubly Cabibbo-suppressed) amplitudes with the mixing of final-state mesons. This effect, estimated to be of an order of $10^{-3}$, is much larger than the direct $CP$ asymmetries in these decays, but missed in the literature. It can be revealed by measuring the difference of the time-dependent $CP$ asymmetries in the $D^{+}\toπ^{+}K_S^0$ and $D_{s}^{+}\to K^{+} K_S^0$ modes, which are accessible at the LHCb and Belle II experiments. If confirmed, the new effect has to be taken into account, as the above direct $CP$ asymmetries are used to search for new physics.

hep-ph↗

arXiv2017

Steering Output Style and Topic in Neural Response Generation

We propose simple and flexible training and decoding methods for influencing output style and topic in neural encoder-decoder based language generation. This capability is desirable in a variety of applications, including conversational systems, where successful agents need to produce language in a specific style and generate responses steered by a human puppeteer or external knowledge. We decompose the neural generation process into empirically easier sub-problems: a faithfulness model and a decoding method based on selective-sampling. We also describe training and sampling algorithms that bias the generation process with a specific language style restriction, or a topic restriction. Human evaluation results show that our proposed methods are able to restrict style and topic without degrading output quality in conversational tasks.

cs.CL↗

arXiv2017

Using Battery Storage for Peak Shaving and Frequency Regulation: Joint Optimization for Superlinear Gains

We consider using a battery storage system simultaneously for peak shaving and frequency regulation through a joint optimization framework which captures battery degradation, operational constraints and uncertainties in customer load and regulation signals. Under this framework, using real data we show the electricity bill of users can be reduced by up to 15\%. Furthermore, we demonstrate that the saving from joint optimization is often larger than the sum of the optimal savings when the battery is used for the two individual applications. A simple threshold real-time algorithm is proposed and achieves this super-linear gain. Compared to prior works that focused on using battery storage systems for single applications, our results suggest that batteries can achieve much larger economic benefits than previously thought if they jointly provide multiple services.

eess.SY↗

arXiv2017

$CP$ violation in charmed hadron decays into neutral kaons

We find a new $CP$ violating effect in charmed hadron decays into neutral kaons, which is induced by the interference between the Cabibbo-favored and doubly Cabibbo-suppressed amplitudes with the $K^{0}-\overline K^{0}$ mixing. It is estimated to be of order of $\mathcal{O}(10^{-3})$, much larger than the direct $CP$ asymmetry, but missed in the literature. To reveal this new $CP$ violation effect, we propose a new observable, the difference of the $CP$ asymmetries in the $D^{+}\to π^{+}K_S^0$ and $D_{s}^{+}\to K^{+} K_S^0$ modes. Once the new effect is determined by experiments, the direct $CP$ asymmetry then can be extracted and used to search for new physics.

hep-ph↗

arXiv2017

Emergence of Topological Nodal Lines and Type II Weyl Nodes in Strong Spin--Orbit Coupling System InNbX2(X=S,Se)

Using first--principles density functional calculations, we systematically investigate electronic structures and topological properties of InNbX2 (X=S, Se). In the absence of spin--orbit coupling (SOC), both compounds show nodal lines protected by mirror symmetry. Including SOC, the Dirac rings in InNbS2 split into two Weyl rings. This unique property is distinguished from other dicovered nodal line materials which normally requires the absence of SOC. On the other hand, SOC breaks the nodal lines in InNbSe2 and the compound becomes a type II Weyl semimetal with 12 Weyl points in the Brillouin Zone. Using a supercell slab calculation we study the dispersion of Fermi arcs surface states in InNbSe2, we also utilize a coherent potential approximation to probe their tolernace to the surface disorder effects. The quasi two--dimensionality and the absence of toxic elements makes these two compounds an ideal experimental platform for investigating novel properties of topological semimetals.

cond-mat.mtrl-sci↗

arXiv2017

First Principles Study of the Giant Magnetic Anisotropy Energy in Bulk Na4IrO4

In 5d transition metal oxides, novel properties arise from the interplay of electron correlations and spin--orbit interactions. Na4IrO4, where 5d transition-metal Ir atom occupies the center of the square-planar coordination environment, is synthesized. Based on density functional theory, we calculate its electronic and magnetic properties. Our numerical results show that the Ir-5d bands are quite narrow, and the bands around the Fermi level are mainly contributed by d_{xy},d_{yz} and d_{zx} orbitals. The magnetic easy-axis is perpendicular to the IrO4 plane, and the magnetic anisotropy energy (MAE) of Na4IrO4 is found to be very giant. We estimate the magnetic parameters by mapping the calculated total energy for different spin configurations onto a spin model. The next nearest neighbor exchange interaction J2 is much larger than other intersite exchange interactions and results in the magnetic ground state configuration. Our study clearly demonstrates that the huge MAE comes from the single-ion anisotropy rather than the anisotropic interatomic spin exchange. This compound has a large spin gap but very narrow spin-wave dispersion, due to the large single-ion anisotropy and relatively small exchange couplings. Noticing this remarkable magnetic feature originated from its highly isolated IrO4 moiety, we also explore the possiblity to further enhance the MAE.

cond-mat.mtrl-sci↗

arXiv2017

$K_{S}^{0}-K_{L}^{0}$ asymmetries in $D$-meson decays

The $K_{S}^{0}-K_{L}^{0}$ asymmetries in the $D$ meson decays, induced by the interference between the Cabibbo-favored and the doubly Cabibbo-suppressed amplitudes, can help to understand the dynamics of charm decays. All possible processes of two-body non-leptonic $D$ decays into one neutral kaon and another pseudoscalar or vector meson are considered. We study the $K_{S}^{0}-K_{L}^{0}$ asymmetries and the branching fractions of corresponding processes in the factorization-assisted topological-amplitude approach in which significant flavor $SU(3)$ symmetry breaking effects are included. The branching fractions of $K_{L}^{0}$ modes are predicted. It is first found that the $K_{S}^{0}-K_{L}^{0}$ asymmetries in the $D^0$-meson decays are shifted by the $D^0-\overline D^0$ mixing parameter $y_{D}\simeq0.006$, to be $0.113\pm0.001$ for all the relevant $D^{0}$ decay modes. Our results on $K_{S}^{0}-K_{L}^{0}$ asymmetries are consistent with the current data and could be tested by experiments in the future.

hep-ph↗

arXiv2016

(CaO)(FeSe): A Layered Wide Gap Oxychalcogenide Semiconductor

A new iron-oxychalcogenide (CaO)(FeSe) was obtained which crystallizes in the orthorhombic space group Pnma (No. 62) with a = 5.9180(12) Å, b = 3.8802(8) Å, c = 13.193(3) Å. The unique structure of (CaO)(FeSe) is built up of a quasi-two-dimensional network of corrugated infinite layers of corner-shared FeSe2O2 tetrahedra that extend in the ab-plane. The corrugated layers composed of corner-shared FeSe2O2 tetrahedra stack along the c-axis with Ca2+ cations sandwiched between the layers. Optical spectroscopy and resistivity measurements reveal semiconducting behavior with an indirect optical band gap of around 1.8 eV and an activation energy of 0.19(1) eV. Electronic band structure calculations at the density function level predict a magnetic configuration as ground state and confirm the presence of an indirect wide gap in (CaO)(FeSe).

cond-mat.mtrl-sci↗

arXiv2016

Leveraging energy storage to optimize data center electricity cost in emerging power markets

Energy storage in data centers has mainly been used as devices to backup generators during power outages. Recently, there has been a growing interest in using energy storage devices to actively shape power consumption in data centers to reduce their skyrocketing electricity bills. In this paper, we consider using energy storage in data centers for two applications in a joint fashion: reducing peak demand charges and enabling data centers to participate in regulation markets. We develop an optimization framework that captures the cost of electricity, degradation of energy storage devices, as well as the benefit from regulation markets. Under this frame- work, using real data Microsoft data center traces and PJM regulation signals, we show the electricity bill of a data center can be reduced by up to 20%. Furthermore, we demonstrate that the saving from joint optimization can be even larger than the sum of individually optimizing each component. We quantify the particular aspects of data center load profiles that lead to this superlinear gain. Compared to prior works that consider using energy storage devices for each single application alone, our results suggest that energy storage in data centers can have much larger impacts than previously thought possible.

cs.DC↗

arXiv2016

CaTe: a new topological node-line and Dirac semimetal

Topological semimetals recently stimulate intense research activities. Combining first-principles calculations and effective model analysis, we predict that CaTe is topological node-line semimetal when spin-orbit coupling (SOC) is ignored. We also obtain the nearly flat surface state which has the drumhead characteristic. When SOC is included, three node lines evolve into a pair of Dirac points along the $M-R$ line. These Dirac points are robust and protected by $C_{4}$ rotation symmetry. Once this crystal symmetry is broken, the Dirac points will be eliminated, and the system becomes a strong topological insulator.

cond-mat.mtrl-sci↗

arXiv2016

Tailoring Magnetic Frustration in Strained Epitaxial FeRh Films

We report on a strain-induced martensitic transformation, accompanied by a suppression of magnetic order in epitaxial films of chemically disordered FeRh. X-ray diffraction, transmission electron microscopy and electronic structure calculations reveal that the lowering of symmetry (from cubic to tetragonal) imposed by the epitaxial relation leads to a further, unexpected, tetragonal-to-orthorhombic transition, triggered by a band-Jahn-Teller-type lattice instability. The collapse of magnetic order is a direct consequence of this structural change, which upsets the subtle balance between ferromagnetic nearest-neighbor interactions arising from Fe-Rh hybridization and frustrated antiferromagnetic coupling among localized Fe moments at larger distances.

cond-mat.mtrl-sci↗

arXiv2015

Faster Parallel Solver for Positive Linear Programs via Dynamically-Bucketed Selective Coordinate Descent

We provide improved parallel approximation algorithms for the important class of packing and covering linear programs. In particular, we present new parallel $ε$-approximate packing and covering solvers which run in $\tilde{O}(1/ε^2)$ expected time, i.e., in expectation they take $\tilde{O}(1/ε^2)$ iterations and they do $\tilde{O}(N/ε^2)$ total work, where $N$ is the size of the constraint matrix and $ε$ is the error parameter, and where the $\tilde{O}$ hides logarithmic factors. To achieve our improvement, we introduce an algorithmic technique of broader interest: dynamically-bucketed selective coordinate descent (DB-SCD). At each step of the iterative optimization algorithm, the DB-SCD method dynamically buckets the coordinates of the gradient into those of roughly equal magnitude, and it updates all the coordinates in one of the buckets. This dynamically-bucketed updating permits us to take steps along several coordinates with similar-sized gradients, thereby permitting more appropriate step sizes at each step of the algorithm. In particular, this technique allows us to use in a straightforward manner the recent analysis from the breakthrough results of Allen-Zhu and Orecchia [2] to achieve our still-further improved bounds. More generally, this method addresses "interference" among coordinates, by which we mean the impact of the update of one coordinate on the gradients of other coordinates. Such interference is a core issue in parallelizing optimization routines that rely on smoothness properties. Since our DB-SCD method reduces interference via updating a selective subset of variables at each iteration, we expect it may also have more general applicability in optimization.

cs.DS↗

arXiv2015

Unified Acceleration Method for Packing and Covering Problems via Diameter Reduction

The linear coupling method was introduced recently by Allen-Zhu and Orecchia for solving convex optimization problems with first order methods, and it provides a conceptually simple way to integrate a gradient descent step and mirror descent step in each iteration. The high-level approach of the linear coupling method is very flexible, and it has shown initial promise by providing improved algorithms for packing and covering linear programs. Somewhat surprisingly, however, while the dependence of the convergence rate on the error parameter $ε$ for packing problems was improved to $O(1/ε)$, which corresponds to what accelerated gradient methods are designed to achieve, the dependence for covering problems was only improved to $O(1/ε^{1.5})$, and even that required a different more complicated algorithm. Given the close connections between packing and covering problems and since previous algorithms for these very related problems have led to the same $ε$ dependence, this discrepancy is surprising, and it leaves open the question of the exact role that the linear coupling is playing in coordinating the complementary gradient and mirror descent step of the algorithm. In this paper, we clarify these issues for linear coupling algorithms for packing and covering linear programs, illustrating that the linear coupling method can lead to improved $O(1/ε)$ dependence for both packing and covering problems in a unified manner, i.e., with the same algorithm and almost identical analysis. Our main technical result is a novel diameter reduction method for covering problems that is of independent interest and that may be useful in applying the accelerated linear coupling method to other combinatorial problems.

cs.DS↗

arXiv2015

Plasmon Resonance in Multilayer Graphene Nanoribbons

Plasmon resonance in nanopatterned single layer graphene nanoribbon (SL-GNR), double layer graphene nanoribbon (DL-GNR) and triple layer graphene nanoribbon (TL-GNR) structures is studied both experimentally and by numerical simulations. We use 'realistic' graphene samples in our experiments to identify the key bottle necks in both experiments and theoretical models. The existence of electrical tunable plasmons in such stacked multilayer GNRs was first experimentally verified by infrared microscopy. We find that the strength of the plasmonic resonance increases in DL-GNR when compared to SL-GNRs. However, we do not find a further such increase in TL-GNRs compared to DL-GNRs. We carried out systematic full wave simulations using finite element technique to validate and fit experimental results, and extract the carrier scattering rate as a fitting parameter. The numerical simulations show remarkable agreement with experiments for unpatterned SLG sheet, and a qualitative agreement for patterned graphene sheet. We believe that further improvements such as introducing a bandgap into the numerical model could lead to a better quantitative agreement of numerical simulations with experiments. We also note that such advanced modeling would first require better quality graphene samples and accurate measurements.

physics.optics↗