arXiv subjects
Publications and source records attributed to Bo Yang.
Motivated by a previous work of Zheng and the second named author, we study pinching constants of compact Kähler manifolds with positive holomorphic sectional curvature. In particular we prove a gap theorem following the work of Petersen and Tao on Riemannian manifolds with almost quarter-pinched sectional curvature.
Given a Hermitian manifold $(M^n,g)$, the Gauduchon connections are the one parameter family of Hermitian connections joining the Chern connection and the Bismut connection. We will call $\nabla^s = (1-\frac{s}{2})\nabla^c + \frac{s}{2}\nabla^b$ the $s$-Gauduchon connection of $M$, where $\nabla^c$ and $\nabla^b$ are respectively the Chern and Bismut connections. It is natural to ask when a compact Hermitian manifold could admit a flat $s$-Gauduchon connection. This is related to a question asked by Yau \cite{Yau}. The cases with $s=0$ (a flat Chern connection) or $s=2$ (a flat Bismut connection) are classified respectively by Boothby \cite{Boothby} in the 1950s or by Q. Wang and the authors recently \cite{WYZ}. In this article, we observe that if either $s\geq 4+2\sqrt{3} \approx 7.46$ or $s\leq 4-2\sqrt{3}\approx 0.54$ and $s\neq 0$, then $g$ is Kähler. We also show that, when $n=2$, $g$ is always Kähler unless $s=2$. Note that non-Kähler compact Bismut flat surfaces are exactly those isosceles Hopf surfaces by \cite{WYZ}.
In this paper, we propose a novel 3D-RecGAN approach, which reconstructs the complete 3D structure of a given object from a single arbitrary depth view using generative adversarial networks. Unlike the existing work which typically requires multiple views of the same object or class labels to recover the full 3D geometry, the proposed 3D-RecGAN only takes the voxel grid representation of a depth view of the object as input, and is able to generate the complete 3D occupancy grid by filling in the occluded/missing regions. The key idea is to combine the generative capabilities of autoencoders and the conditional Generative Adversarial Networks (GAN) framework, to infer accurate and fine-grained 3D structures of objects in high-dimensional voxel space. Extensive experiments on large synthetic datasets show that the proposed 3D-RecGAN significantly outperforms the state of the art in single view 3D object reconstruction, and is able to reconstruct unseen types of objects. Our code and data are available at: https://github.com/Yang7879/3D-RecGAN.
An important question for the practical applicability of the highly efficient traffic intersection control is about the minimal level of intelligence the vehicles need to have so as to move beyond the traffic light control. We propose an efficient intersection traffic control scheme without the traffic lights, that only requires a majority of vehicles on the road to be equipped with a simple driver assistance system. The algorithm of our scheme is completely decentralized, and takes into full account the non-linear interaction between the vehicles at high density. For vehicles approaching the intersection in different directions, our algorithm imposes simple interactions between vehicles around the intersection, by defining specific conditions on the real-time basis, for which the involved vehicles are required to briefly adjust their dynamics. This leads to a self-organised traffic flow that is safe, robust, and efficient. We also take into account of the driver comfort level and study its effect on the control efficiency. The scheme has low technological barrier, minimal impact on the conventional driving behaviour, and can coexist with the traffic light control. It also has the advantages of being easily scalable, and fully compatible with both the conventional road systems as well as the futuristic scenario in which driverless vehicles dominate the road. The mathematical formulation of our scheme permits large scale realistic numerical simulations of busy intersections, allowing a more complete evaluation of the control performance, instead of just the collision avoidance at the intersection.
The goal of load balancing (grid partitioning) is to minimize overall computations and communications, and to make sure that all processors have a similar workload. Geometric methods divide a grid by using a location of a cell while topological methods work with connectivity of cells, which is generally described as a graph. This paper introduces a Hilbert space-filling curve method. A space-filling curve is a continuous curve and defines a map between a one-dimensional space and a multi-dimensional space. A Hilbert space-filling curve is one special space-filling curve discovered by Hilbert and has many useful characteristics, such as good locality, which means that two objects that are close to each other in a multi-dimensional space are also close to each other in a one dimensional space. This property can model communications in grid-based parallel applications. The idea of the Hilbert space-filling curve method is to map a computational domain into a one-dimensional space, partition the one-dimensional space to certain intervals, and assign all cells in a same interval to a MPI. To implement a load balancing method, a mapping kernel is required to convert high-dimensional coordinates to a scalar value and an efficient one-dimensional partitioning module that divides a one-dimensional space and makes sure that all intervals have a similar workload. The Hilbert space-filling curve method is compared with ParMETIS, a famous graph partitioning package. The results show that our Hilbert space-filling curve method has good partition quality. It has been applied to grids with billions of cells, and linear scalability has been obtained on IBM Blue Gene/Q.
In this paper, the influence of the fractional dimensions of the Lévy path under the Earth's gravitational field is studied, and the phase transitions of energy and wave functions are obtained: the energy changes from discrete to continuous and wave functions change from non-degenerate to degenerate when dimension of Lévy path becomes from integer to non-integer. By analyzing the phase transitions, we solve two popular problems. First, we find an exotic way to produce the bound states in the continuum (BICs), our approach only needs a simple potential, and does not depend on interactions between particles. Second, we address the continuity of the energy will become strong when the mass of the particle becomes small. By deeply analyze, it can provide a way to distinguish ultralight particles from others types in the Earth's gravitational field, and five popular particles are discussed. In addition, we obtain analytical expressions for the wave functions and energy in the Earth's gravitational field in the circumstance of a fractional fractal dimensional Lévy path. Moreover, to consider the influence of the minimal length, we analyze the phase transitions and the BICs in the presence of the minimal length. We find the phenomenon energy shift do not exist, which is a common phenomenon in the presence of the minimal length, and hence such above phenomena can still be found. Finally, relations between our results and existing results are discussed.
Most learning approaches treat dimensionality reduction (DR) and clustering separately (i.e., sequentially), but recent research has shown that optimizing the two tasks jointly can substantially improve the performance of both. The premise behind the latter genre is that the data samples are obtained via linear transformation of latent representations that are easy to cluster; but in practice, the transformation from the latent space to the data can be more complicated. In this work, we assume that this transformation is an unknown and possibly nonlinear function. To recover the `clustering-friendly' latent representations and to better cluster the data, we propose a joint DR and K-means clustering approach in which DR is accomplished via learning a deep neural network (DNN). The motivation is to keep the advantages of jointly optimizing the two tasks, while exploiting the deep neural network's ability to approximate any nonlinear function. This way, the proposed approach can work well for a broad class of generative models. Towards this end, we carefully design the DNN structure and the associated joint optimization criterion, and propose an effective and scalable algorithm to handle the formulated optimization problem. Experiments using different real datasets are employed to showcase the effectiveness of the proposed approach.
We investigate the dynamics of the rate function and of local observables after a quench in models which exhibit phase transitions between a superfluid and an insulator in their ground states. Zeros of the return probability, corresponding to singularities of the rate functions, have been suggested to indicate the emergence of dynamical criticality and we address the question of whether such zeros can be tied to the dynamics of physically relevant observables and hence order parameters in the systems. For this we first numerically analyze the dynamics of a hard-core boson gas in a one-dimensional waveguide when a quenched lattice potential is commensurate with the particle density. Such a system can undergo a pinning transition to an insulating state and we find non-analytic behavior in the evolution of the rate function which is indicative of dynamical phase transitions. In addition, we perform simulations of the time dependence of the momentum distribution and compare the periodicity of this collapse and revival cycle to that of the non-analyticities in the rate function: the two are found to be closely related only for deep quenches. We then confirm this observation by analytic calculations on a closely related discrete model of hard-core bosons in the presence of a staggered potential and find expressions for the rate function for the quenches. By extraction of the zeros of the Loschmidt amplitude we uncover a non-equilibrium timescale for the emergence of non-analyticities and discuss its relationship with the dynamics of the experimentally relevant parity operator.
Recently, a number of nonlocal integrable equations, such as the PT-symmetric nonlinear Schrodinger (NLS) equation and PT-symmetric Davey-Stewartson equations, were proposed and studied. Here we show that many of such nonlocal integrable equations can be converted to local integrable equations through simple variable transformations. Examples include these nonlocal NLS and Davey-Stewartson equations, a nonlocal derivative NLS equation, the reverse space-time complex modified Korteweg-de Vries (CMKdV) equation, and many others. These transformations not only establish immediately the integrability of these nonlocal equations, but also allow us to construct their analytical solutions from solutions of the local equations. These transformations can also be used to derive new nonlocal integrable equations. As applications of these transformations, we use them to derive rogue wave solutions for the partially PT-symmetric Davey-Stewartson equations and the nonlocal derivative NLS equation. In addition, we use them to derive multi-soliton and quasi-periodic solutions in the reverse space-time CMKdV equation. Furthermore, we use them to construct many new nonlocal integrable equations such as nonlocal short pulse equations, nonlocal nonlinear diffusion equations, and nonlocal Sasa-Satsuma equations.
We generalize the notion of Haldane pseudopotentials to anisotropic fractional quantum Hall (FQH) systems which are physically realized, e.g., in tilted magnetic field experiments or anisotropic band structures. This formalism allows us to expand any translation-invariant interaction over a complete basis, and directly reveals the intrinsic metric of incompressible FQH fluids. We show that purely anisotropic pseudopotentials give rise to new types of bound states for small particle clusters in the infinite plane, and can be used as a diagnostic of FQH nematic order. We also demonstrate that generalized pseudopotentials quantify the anisotropic contribution to the effective interaction potential, which can be particularly large in models of fractional Chern insulators.
High penetration of renewable energy source makes microgrid (MGs) be environment friendly. However, the stochastic input from renewable energy resource brings difficulty in balancing the energy supply and demand. Purchasing extra energy from macrogrid to deal with energy shortage will increase MG energy cost. To mitigate intermittent nature of renewable energy, energy trading and energy storage which can exploit diversity of renewable energy generation across space and time are efficient and cost-effective methods. But current energy storage control action will impact the future control action which brings challenge to energy management. In addition, due to MG participating energy trading as prosumer, it calls for an efficient trading mechanism. Therefore, this paper focuses on the problem of MG energy management and trading. Energy trading problem is formulated as a stochastic optimization one with both individual profit and social welfare maximization. Firstly a Lyapunov optimization based algorithm is developed to solve the stochastic problem. Secondly the double-auction based mechanism is provided to attract MG truthful bidding for buying and selling energy. Through theoretical analysis, we demonstrate that individual MG can achieve a time average energy cost close to offline optimum with tradeoff between storage capacity and energy trading cost. Meanwhile the social welfare is also asymptotically maximized under double auction. Simulation results based on real world data show the effectiveness of our algorithm.
This research investigates the implementation mechanism of block-wise ILU(k) preconditioner on GPU. The block-wise ILU(k) algorithm requires both the level k and the block size to be designed as variables. A decoupled ILU(k) algorithm consists of a symbolic phase and a factorization phase. In the symbolic phase, a ILU(k) nonzero pattern is established from the point-wise structure extracted from a block-wise matrix. In the factorization phase, the block-wise matrix with a variable block size is factorized into a block lower triangular matrix and a block upper triangular matrix. And a further diagonal factorization is required to perform on the block upper triangular matrix for adapting a parallel triangular solver on GPU.We also present the numerical experiments to study the preconditioner actions on different k levels and block sizes.
In urban transportation systems, mobility flows in the subway system reflect the spatial and temporal dynamics of working days. To investigate the variability of mobility flows, we analyse the spatial community through a series of snapshots of subway stations over sequential periods. Using Shanghai as a case study, we find that the spatial community snapshots reveal dynamic passenger activities. Adopting a dual-perspective, we apply spatial and temporal models separately to explore where and when individuals travel for entertainment. In the two models, microblog topics and spatial facilities such as food venues and entertainment businesses are used to characterise the spatial popularity of each station and people's travelling perceptions. In the studied case, the city centre is characterised by greater social influence, and it is better described by the spatial model. In the temporal model, shorter travel distances motivate individuals to start their trips earlier. Interestingly, as the number of food-related facilities near the starting station increases, until it exceeds 1563, the speed of people's journeys slows down. This study provides a method for modelling the effects of social features on mobility flows and for predicting the spatial-temporal mobility flows of newly built subway stations.
This paper presents our work on developing parallel computational methods for two-phase flow on modern parallel computers, where techniques for linear solvers and nonlinear methods are studied and the standard and inexact Newton methods are investigated. A multi-stage preconditioner for two-phase flow is applied and advanced matrix processing strategies are studied. A local reordering method is developed to speed the solution of linear systems. Numerical experiments show that these computational methods are effective and scalable, and are capable of computing large-scale reservoir simulation problems using thousands of CPU cores on parallel computers. The nonlinear techniques, preconditioner and matrix processing strategies can also be applied to three-phase black oil, compositional and thermal models.
As a new type of quantum matter, Dirac node line (DNL) semimetals are currently attracting widespread interest in condensed matter physics and material science. The DNL featured by a closed line consisting of linear band crossings in the lattice momentum space are mostly predicted in three-dimensional materials. Here, we propose tight-binding (TB) models of pz/px,y or pz/s orbitals in a two-dimensional (2D) bipartite square lattice for the 2D version of DNL semimetals. The DNL states in these models are caused by the inversion of the bands with different symmetries and thus robust again spin-orbit coupling (SOC). By means of first-principles calculations, we demonstrate two candidate 2D materials of these models: Be2C and BeH2 monolayers, which have Fermi circles centered at Γ (0,0) and K (1/2, 1/2) points, respectively. The topological nontriviality is verified by the non-zero topological invariant and the edge states. This work opens an avenue for design of 2D DNL semimetals.
This paper is the first step in a systematic project to study examples of Kähler manifolds with positive holomorphic sectional curvature ($H > 0$). Previously Hitchin proved that any compact Kähler surface with $H>0$ must be rational and he constructed such examples on Hirzebruch surfaces $M_{2, k}=\mathbb{P}(H^{k}\oplus 1_{\mathbb{CP}^1})$. We generalize Hitchin's construction and prove that any Hirzebruch manifold $M_{n, k}=\mathbb{P}(H^{k}\oplus 1_{\mathbb{CP}^{n-1}})$ admits a Kähler metric of $H>0$ in each of its Kähler classes. We demonstrate that the pinching behaviors of holomorphic sectional curvatures of new examples differ from those of Hitchin's which were studied in the recent work of Alvarez-Chaturvedi-Heier. Some connections to recent works on the Kähler-Ricci flow on Hirzebruch manifolds are also discussed. It seems interesting to study the space of all Kähler metrics of $H>0$ on a given Kähler manifold. We give higher dimensional examples such that some Kähler classes admit Kähler metrics with $H>0$ and some do not.
A parallel reservoir simulator has been developed, which is designed for large-scale black oil simulations. It handles three phases, including water, oil and gas, and three components, including water, oil and gas. This simulator can calculate traditional reservoir models and naturally fractured models. Various well operations are supported, such as water flooding, gas flooding and polymer flooding. The operation constraints can be fixed bottom-hole pressure, a fixed fluid rate, and combinations of them. The simulator is based on our in-house platform, which provides grids, cell-centred data, linear solvers, preconditioners and well modeling. The simulator and the platform use MPI for communications among computation nodes. Our simulator is capable of simulating giant reservoir models with hundreds of millions of grid cells. Numerical simulations show that our simulator matches with commercial simulators and it has excellent scalability.
Doping has been used to enhance the ionization efficiency of analytes in atmospheric pressure photoionization, which is based on charge exchange. Compounds with excellent ionization efficiencies are usually chosen as dopants. In this paper, we report a new phenomenon observed in low-pressure photoionization: Protonation enhancement by dichloromethane (CH2Cl2) doping. CH2Cl2 is not a common dopant due to its high ionization energy (11.33 eV). The low-pressure photoionization source was built using a krypton VUV lamp that emits photons with energies of 10.0 and 10.6 eV and was operated at 500-1000 Pa. Protonation of water, methanol, ethanol, and acetaldehyde was respectively enhanced by 481.7 +/- 122.4, 197.8 +/- 18.8, 87.3 +/- 7.8, and 93.5 +/- 35.5 times after doping 291 ppmv CH2Cl2, meanwhile CH2Cl2 almost does not generate noticeable ions itself. This phenomenon has not been documented in the literature. A new protonation process involving in ion-pair and H-bond formations was proposed to expound the phenomenon. The observed phenomenon opens a new prospect for the improvement of the detection efficiency of VUV photoionization.