Results for "Haotian Chi"
total 2812took 0.15s
Joint Stochastic Approximation learning of Helmholtz MachinesMar 20 2016Sep 28 2018Though with progress, model learning and performing posterior inference still remains a common challenge for using deep generative models, especially for handling discrete hidden variables. This paper is mainly concerned with algorithms for learning Helmholz ... More On type-II singularities in Ricci flow on $\mathbb{R}^{N}$Oct 15 2012Sep 19 2015In each dimension $N\geq 3$ and for each real number $\lambda\geq 1$, we construct a family of complete rotationally symmetric solutions to Ricci flow on $\mathbb{R}^{N}$ which encounter a global singularity at a finite time $T$. The singularity forms ... More Two Stage Optimization with Recourse and RevocationOct 21 2016Two-stage optimization with recourse model is an important and widely used model, which has been studied extensively these years. In this article, we will look at a new variant of it, called the two-stage optimization with recourse and revocation model. ... More Point-Voxel CNN for Efficient 3D Deep LearningJul 08 2019We present Point-Voxel CNN (PVCNN) for efficient, fast 3D deep learning. Previous work processes 3D data using either voxel-based or point-based NN models. However, both approaches are computationally inefficient. The computation cost and memory footprints ... More K-semistability is equivariant volume minimizationDec 22 2015Sep 27 2016This is a continuation to the paper \cite{Li15a} in which a problem of minimizing normalized volumes over $\mathbb{Q}$-Gorenstein klt singularities was proposed. Here we consider the case when there is a $\mathbb{C}^*$-action on $(X,o)$. In particular, ... More False discovery rate control with multivariate $p$-valuesJun 04 2007May 21 2008Multivariate statistics are often available as well as necessary in hypothesis tests. We study how to use such statistics to control not only false discovery rate (FDR) but also positive FDR (pFDR) with good power. We show that FDR can be controlled through ... More Two-phase flow of ferrofluidsJul 26 2016Ferrofluids currently are the only type of magnetic liquid materials with wide practical use. The theory on ferrofluids is an example of success to apply statistics to science. Ferrofluids are two-phase liquids consisting of dispersed nanoscale ferromagnetic ... More QCD strings and the U(1) problemNov 30 2018We promote the usual QCD $\theta$-parameter to a field and interpret it as the phase of the quark condensate, which becomes nontrivial when topological defects, vortices in our formulation, are induced in the quark condensate by the QCD strings (chromoelectric ... More On Multivariate Strong Renewal TheoremDec 05 2014Mar 15 2017This paper takes the so-called probabilistic approach to the Strong Renewal Theorem (SRT) for multivariate distributions in the domain of attraction of a stable law. A version of the SRT is obtained that allows any kind of lattice-nonlattice composition ... More Large deviations for template matching between point processesMar 22 2005We study the asymptotics related to the following matching criteria for two independent realizations of point processes X\sim X and Y\sim Y. Given l>0, X\cap [0,l) serves as a template. For each t>0, the matching score between the template and Y\cap [t,t+l) ... More Gemini: Reducing DRAM Cache Hit Latency by Hybrid MappingsJun 03 2018Die-stacked DRAM caches are increasingly advocated to bridge the performance gap between on-chip Cache and main memory. It is essential to improve DRAM cache hit rate and lower cache hit latency simultaneously. Prior DRAM cache designs fall into two categories ... More On Multivariate Strong Renewal TheoremDec 05 2014Dec 09 2014This paper takes the so-called probabilistic approach to the Strong Renewal Theorem (SRT) on $\mathbb{R}^d$ for distributions in the domain of attraction without centering of a nondegenerate stable law of exponent $\alpha \in (0, 2]\cap (0,d)$. We obtain ... More Subspace Learning From BitsJul 23 2014Jul 24 2014This paper proposes a simple sensing and estimation framework to faithfully recover the principal subspace of high-dimensional datasets or data streams from a collection of one-bit measurements from distributed sensors based on comparing accumulated energy ... More A de-gauging approach to physics beyond the Standard ModelJun 06 2016By studying the t-J model for superconductivity, the Pati-Salam model and the Haplon model for particle unifications, we extract their common feature which is the spin-charge separation of fermions. This becomes a de-gauging process for charged fermions ... More Phase of Quark Condensate and Topological Current in QCDOct 31 2013We propose a new topological charge term in QCD based on flux-tube models. It couples a superflow of the phase of the quark condensate to the Chern-Simons current. The usual $\theta$-parameter is replaced by the phase of the quark condensate, which becomes ... More Generative Models for Learning from CrowdsJun 13 2017Oct 03 2017In this paper, we propose generative probabilistic models for label aggregation. We use Gibbs sampling and a novel variational inference algorithm to perform the posterior inference. Empirical results show that our methods consistently outperform state-of-the-art ... More Joint Sparsity Recovery for Spectral Compressed SensingNov 09 2013Feb 03 2014Compressed Sensing (CS) is an effective approach to reduce the required number of samples for reconstructing a sparse signal in an a priori basis, but may suffer severely from the issue of basis mismatch. In this paper we study the problem of simultaneously ... More Lorentz Violation by Quark CondensationAug 02 2010At the TeV scale, heavy quarks, for example the 4th generation quarks of the Standard Model with four generations, can form condensates which dynamically break the electroweak symmetry. A dense quark system may form other types of condensates which dynamically ... More Geometry of Kottwitz-Viehmann VarietiesMay 22 2018We study basic geometric properties of Kottwitz-Viehmann varieties, which are certain generalizations of affine Springer fibers that encode orbital integrals of spherical Hecke functions. Based on previous work of A. Bouthier and the author, we show that ... More Molecular Clock Dating using MrBayesMar 17 2016There are two modern approaches on dating species divergence using molecular data: node dating (e.g., Yang and Rannala, 2006; Drummond et al., 2006) and total-evidence dating (e.g., Ronquist et al., 2012a; Zhang et al., 2016). In a Bayesian framework, ... More Minimizing normalized volumes of valuationsNov 25 2015Apr 10 2016We propose a problem of minimizing normalized volumes of valuations. For a $\mathbb{Q}$-Gorenstein klt singularity, by proving a key properness estimate, we prove that the set of valuations with uniformly bounded normalized volumes is compact. Under a ... More Gauged Nambu-Jona-Lasinio model and axionic QCD stringDec 30 2014We propose an axionic QCD string scenario based on the original flux-tube model by Kogut and Susskind, and then incorporate it into a gauged Nambu-Jona-Lasinio (NJL) model. Axial anomaly is studied by a new topological coupling from the string side, and ... More QCD Flux Tubes and Anomaly InflowFeb 28 2013Jul 17 2013We apply the Callan-Harvey anomaly inflow mechanism to the study of QCD (chromoelectric) flux tubes, quark (pair)-creation and chiral magnetic effect, using new variables from the Cho-Faddeev-Niemi decomposition of the gauge potential. A phenomenological ... More The geometry of some generalized affine Springer fibersOct 30 2017May 23 2018We study basic geometric properties of some group analogue of affine Springer fibers and compare with the classical Lie algebra affine Springer fibers. The main purpose is to formulate a conjecture that relates the number of irreducible components of ... More Joint Stochastic Approximation learning of Helmholtz MachinesMar 20 2016Though with progress, model learning and performing posterior inference still remains a common challenge for using deep generative models, especially for handling discrete hidden variables. This paper is mainly concerned with algorithms for learning Helmholz ... More Evolving Agents for the Hanabi 2018 CIG CompetitionSep 26 2018Hanabi is a cooperative card game with hidden information that has won important awards in the industry and received some recent academic attention. A two-track competition of agents for the game will take place in the 2018 CIG conference. In this paper, ... More A Model-based Approach for Effective Service DeliveryApr 19 2016May 31 2016With the prevalence of X-as-a-Service (e.g., software as a service, platform as a service, infrastructure as a service, etc.) and users' growing demand on good services, QoS (Quality of Service) assurance is becoming increasingly important to service ... More Berkovich spectra of elements in Banach RingsOct 22 2014Adapting the notion of the spectrum $\Sigma_a$ for an element $a$ in an ultrametric Banach algebra (as defined by Berkovich), we introduce and briefly study the Berkovich spectrum $\sigma^{Ber}_R(u)$ of an element $u$ in a Banach ring $R$. This spectrum ... More Dynamical stability of algebraic Ricci solitonsSep 21 2013Sep 10 2014We consider dynamical stability for a modified Ricci flow equation whose stationary solutions include Einstein and Ricci soliton metrics. Our focus is on homogeneous metrics on non-compact manifolds. Following the program of Guenther, Isenberg, and Knopf, ... More Conical Kahler-Einstein metric revisitedJul 20 2012Oct 07 2012In this paper we introduce the "interpolation-degneration" strategy to study Kahler-Einstein metrics on a smooth Fano manifold with cone singularities along a smooth divisor that is proportional to the anti-canonical divisor. By "interpolation" we show ... More Geometric Bijections Between Spanning Trees and Break DivisorsSep 09 2015The Jacobian group ${\rm Jac}(G)$ of a finite graph $G$ is a group whose cardinality is the number of spanning trees of $G$. $G$ also has a tropical Jacobian which has the structure of a real torus; using the notion of break divisors, An et al. obtained ... More Kähler-Einstein metrics and volume minimizationFeb 16 2016Apr 11 2016We prove that if a $\mathbb{Q}$-Fano variety $V$ specially degenerates to a K\"ahler-Einstein $\mathbb{Q}$-Fano variety $V$, then for any ample Cartier divisor $H=-r^{-1} K_V$ with $r\in \mathbb{Q}_{>0}$, the normalized volume $\widehat{\rm vol}(v)=A_X^n(v)\cdot ... More Some remarks on groupoids and small categoriesOct 18 2007This unpublished note contains some materials taken from my old study note on groupoids and small categories. It contains a proof for the fact that any groupoid is a group bundle over an equivalence relation. Moreover, the action of a category $G$ on ... More Domains of Injectivity for the Gross-Hopkins Period MapNov 29 2013Mar 11 2015We determine the domain of injectivity of the Gross-Hopkins Period map around each points in the deformation space for a fixed formal module $\bar{F}$ of height 2 that defined over a finite field. And then we will use this to conclude some local analyticity ... More A new look at Condition AJul 02 2009Ozeki and Takeuchi \cite[I]{OT} introduced the notion of Condition A and Condition B to construct two classes of inhomogeneous isoparametric hypersurfaces with four principal curvatures in spheres, which were later generalized by Ferus, Karcher and M\"unzner ... More Universal programmable devices for unambiguous discriminationJun 22 2006We discuss the problem of designing unambiguous programmable discriminators for any $n$ unknown quantum states in an $m$-dimensional Hilbert space. The discriminator is a fixed measurement which has two kinds of input registers: the program registers ... More Duality of Hopf $C^*$-algebrasJun 03 2002In this paper, we study the duality theory of Hopf $C^*$-algebras in a general ``Hilbert-space-free'' framework. Our particular interests are the ``full duality'' and the ``reduced duality''. In order to study the reduced duality, we define the interesting ... More Stochastic Fractional HP EquationsJun 24 2009In this paper we established the condition for a curve to satisfy stochas- tic fractional HP (Hamilton-Pontryagin) equations. These equations are described using It^o integral. We have also considered the case of stochastic fractional Hamiltonian equa- ... More Microwave-mediated heat transport through a quantum dotFeb 06 2012The thermoelectric effect in a quantum dot (QD) attached to two leads in the presence of microwave fields is studied by using the Keldysh nonequilibrium Green function technique. When the microwave is applied only on the QD and in the linear-response ... More Isoparametric hypersurfaces with four principal curvatures, IVMay 03 2016We prove that an isoparametric hypersurface with four principal curvatures and multiplicity pair $(7,8)$ is either the one constructed by Ozeki and Takeuchi, or one of the two constructed by Ferus, Karcher, and M\"{u}nzner. This completes the classification ... More Generalized Cluster Trees and Singular MeasuresNov 08 2016In this paper, we study the $\alpha$-cluster tree ($\alpha$-tree) under both singular and nonsingular measures. The $\alpha$-tree uses probability contents within a level set to construct a cluster tree so that it is well-defined for singular measures. ... More Do recommender systems benefit users?Jul 05 2015Recommender systems are present in many web applications to guide our choices. They increase sales and benefit sellers, but whether they benefit customers by providing relevant products is questionable. Here we introduce a model to examine the benefit ... More Graviton Bending in Quantum Gravity from One-Loop AmplitudesMar 19 2019We study the bending of gravitons that pass near a massive object like the Sun, using scattering amplitudes in which the Sun is represented by a massive scalar particle. Our results complete previous work on the bending angles of massless spin-0, spin-$\frac{1}{2}$ ... More