arXiv subjects

Kai Li

Publications and source records attributed to Kai Li.

At least 451 records · Page 25Linked to original sources

arXiv2015

From rules to runs: A dynamic epistemic take on imperfect information games

In the literature of game theory, the information sets of extensive form games have different interpretations, which may lead to confusions and paradoxical cases. We argue that the problem lies in the mix-up of two interpretations of the extensive form game structures: game rules or game runs which do not always coincide. In this paper, we try to separate and connect these two views by proposing a dynamic epistemic framework in which we can compute the runs step by step from the game rules plus the given assumptions of the players. We propose a modal logic to describe players' knowledge and its change during the plays, and provide a complete axiomatization. We also show that, under certain conditions, the mix-up of the rules and the runs is not harmful due to the structural similarity of the two.

cs.AI↗

arXiv2015

Numerical Modeling on Thermal Loading of Diamond Crystal in X-ray FEL Oscillator

Due to high reflectivity and high resolution of X-ray pulses, diamond is one of the most popular Bragg crystals serving as the reflecting mirror and mono$-$chromator in the next generation of free electrons lasers (FELs). The energy deposition of X-rays will result in thermal heating, and thus lattice expansion of the diamond crystal, which may degrade the performance of X-ray FELs. In this paper, the thermal loading effect of diamond crystal for X-ray FEL oscillators has been systematically studied by combined simulation with Geant4 and ANSYS, and its dependence on the environmental temperature, crystal size, X-ray pulse repetition rate and pulse energy are presented. Our results show that taking the thermal loading effects into account, X-ray FEL oscillators are still robust and promising with an optimized design.

physics.acc-ph↗

arXiv2015

First-Take-All: Temporal Order-Preserving Hashing for 3D Action Videos

With the prevalence of the commodity depth cameras, the new paradigm of user interfaces based on 3D motion capturing and recognition have dramatically changed the way of interactions between human and computers. Human action recognition, as one of the key components in these devices, plays an important role to guarantee the quality of user experience. Although the model-driven methods have achieved huge success, they cannot provide a scalable solution for efficiently storing, retrieving and recognizing actions in the large-scale applications. These models are also vulnerable to the temporal translation and warping, as well as the variations in motion scales and execution rates. To address these challenges, we propose to treat the 3D human action recognition as a video-level hashing problem and propose a novel First-Take-All (FTA) Hashing algorithm capable of hashing the entire video into hash codes of fixed length. We demonstrate that this FTA algorithm produces a compact representation of the video invariant to the above mentioned variations, through which action recognition can be solved by an efficient nearest neighbor search by the Hamming distance between the FTA hash codes. Experiments on the public 3D human action datasets shows that the FTA algorithm can reach a recognition accuracy higher than 80%, with about 15 bits per frame considering there are 65 frames per video over the datasets.

cs.CV↗

arXiv2015

Enhanced trapping of colding lithium by using the multiple-sideband cooling in a two-dimensional magneto-optical trap

Trapping lithium with a big number in a simplified experimental setup has difficulties and challenges today. In this paper, we experimentally demonstrate the enhancement of \textsuperscript{6}Li trapping efficiency in a three-dimensional magneto-optical trap (3D MOT) by using the multiple-sideband cooling in a two-dimensional magneto-optical trap (2D MOT). To improve the number of trapped atoms, we broaden the cooling light spectrum to 102 MHz composed of seven frequency components and then trap atoms with a number of $6.0\times10^8$ which is about 4 times compared to that in the single-frequency cooling. The capture velocity and dependence of atomic number on the laser detuning have been analyzed, where the experimental result has a good agreement with the theoretical prediction based on a simple two-level model. We also analyze the loss rate of alkali metals due to fine-structure exchanging collisions and find that the multiple-sideband cooling is special valid for lithium.

physics.atom-ph↗

arXiv2015

Global phase diagram, possible chiral spin liquid and topological superconductivity in the triangular Kitaev-Heisenberg model

The possible ground states of the undoped and doped Kitaev-Heisenberg model on a triangular lattice are studied. For the undoped system, a combination of the numerical exact diagonalization calculation and the four-sublattice transformation analysis suggests one possible exotic phase and four magnetically ordered phases including a collinear stripe pattern and a noncollinear spiral pattern in the global phase diagram. The exotic phase near the antiferromagnetic (AF) Kitaev point is further investigated by using the Schwinger-fermion mean-field method, and we obtain an energetically favorable $Z_2$ chiral spin liquid with a Chern number $\pm2$ as a promising candidate. At finite doping, we find that the AF Heisenberg coupling supports an $s$-wave or a $d_{x^2-y^2}+id_{xy}$-wave superconductivity (SC), while the AF and the ferromagnetic Kitaev interactions favor a $d_{x^2-y^2}+id_{xy}$-wave SC and a time-reversal invariant topological $p$-wave SC, respectively. Possible experimental realizations and related candidate materials are also discussed.

cond-mat.str-el↗

arXiv2015

Rank Subspace Learning for Compact Hash Codes

The era of Big Data has spawned unprecedented interests in developing hashing algorithms for efficient storage and fast nearest neighbor search. Most existing work learn hash functions that are numeric quantizations of feature values in projected feature space. In this work, we propose a novel hash learning framework that encodes feature's rank orders instead of numeric values in a number of optimal low-dimensional ranking subspaces. We formulate the ranking subspace learning problem as the optimization of a piece-wise linear convex-concave function and present two versions of our algorithm: one with independent optimization of each hash bit and the other exploiting a sequential learning framework. Our work is a generalization of the Winner-Take-All (WTA) hash family and naturally enjoys all the numeric stability benefits of rank correlation measures while being optimized to achieve high precision at very short code length. We compare with several state-of-the-art hashing algorithms in both supervised and unsupervised domain, showing superior performance in a number of data sets.

cs.LG↗

arXiv2015

GRB Spectra in the complex of synchrotron and Compton processes

Under the steady state condition, the spectrum of electrons is investigated by solving the continuity equation under the complex radiation of both the synchrotron and Compton processes. The resulted GRB spectrum is a broken power law in both the fast and slow cooling phases. On the basis of this electron spectrum, the spectral indices of the Band function in four different phases are presented. In the complex radiation frame, the detail investigation on physical parameters reveals that both the reverse shock photosphere model and the forward shock with strong coupling model can answer the $α\sim -1$ problem. A possible marginal to fast cooling phase transition in GRB 080916C is discussed. The time resolved spectra in different pulses of GRB 100724B, GRB 100826A and GRB 130606B are investigated. We found that the flux is proportional to the peak energy in almost all pulses. The phases for different pulses are determined according to the spectral index revolution. We found the strong correlations between spectral indices and the peak energy in GRB 100826A, which can be explained by the Compton effect in the fast cooling phase. However, the complex scenario predicts a steeper index for the injected electrons, which challenges the acceleration mechanism in GRBs.

astro-ph.HE↗

arXiv2014

Electromagnetially-induced-transparency-like ground-state cooling in a double-cavity optomechanical system

We propose to cool a mechanical resonator close to its ground state via an electromagnetically-induced-transparency- (EIT-) like cooling mechanism in a double-cavity optomechanical system, where an additional cavity couples to the original one in the standard optomechanical system. By choosing optimal parameters such that the cooling process of the mechanical resonator corresponds to the maximum value of the optical fluctuation spectrum and the heating process to the minimum one, the mechanical resonator can be cooled with the final mean phonon number less than that at the absence of the additional cavity. And we show the mechanical resonator may be cooled close to its ground state via such an EIT-like cooling mechanism even when the original resolved sideband condition is not fulfilled at the absence of the additional cavity.

quant-ph↗

arXiv2014

A Non-Cooperative Game Model for Reliability-Based Task Scheduling in Cloud Computing

Cloud computing is a newly emerging distributed system which is evolved from Grid computing. Task scheduling is the core research of cloud computing which studies how to allocate the tasks among the physical nodes, so that the tasks can get a balanced allocation or each task's execution cost decreases to the minimum, or the overall system performance is optimal. Unlike task scheduling based on time or cost before, aiming at the special reliability requirements in cloud computing, we propose a non-cooperative game model for reliability-based task scheduling approach. This model takes the steady-state availability that computing nodes provide as the target, takes the task slicing strategy of the schedulers as the game strategy, then finds the Nash equilibrium solution. And also, we design a task scheduling algorithm based on this model. The experiments can be seen that our task scheduling algorithm is better than the so-called balanced scheduling algorithm.

cs.DC↗

arXiv2014

A Task Allocation Schema Based on Response Time Optimization in Cloud Computing

Cloud computing is a newly emerging distributed computing which is evolved from Grid computing. Task scheduling is the core research of cloud computing which studies how to allocate the tasks among the physical nodes so that the tasks can get a balanced allocation or each task's execution cost decreases to the minimum or the overall system performance is optimal. Unlike the previous task slices' sequential execution of an independent task in the model of which the target is processing time, we build a model that targets at the response time, in which the task slices are executed in parallel. Then we give its solution with a method based on an improved adjusting entropy function. At last, we design a new task scheduling algorithm. Experimental results show that the response time of our proposed algorithm is much lower than the game-theoretic algorithm and balanced scheduling algorithm and compared with the balanced scheduling algorithm, game-theoretic algorithm is not necessarily superior in parallel although its objective function value is better.

cs.DC↗

arXiv2014

Optimal Switching Problems under Partial Information

In this paper we formulate and study an optimal switching problem under partial information. In our model the agent/manager/investor attempts to maximize the expected reward by switching between different states/investments. However, he is not fully aware of his environment and only an observation process, which contains partial information about the environment/underlying, is accessible. It is based on the partial information carried by this observation process that all decisions must be made. We propose a probabilistic numerical algorithm based on dynamic programming, regression Monte Carlo methods, and stochastic filtering theory to compute the value function. In this paper, the approximation of the value function and the corresponding convergence result are obtained when the underlying and observation processes satisfy the linear Kalman-Bucy setting. A numerical example is included to show some specific features of partial information.

math.OC↗

arXiv2014

A central limit type theorem for Gaussian mixture approximations to the nonlinear filtering problem

Approximating the solution of the nonlinear filtering problem with Gaussian mixtures has been a very popular method since the 1970s. However, the vast majority of such approximations are introduced in an ad-hoc manner without theoretical grounding. This work is a continuation of [4, 5], where we described a rigorous Gaussian mixture approximation to the solution of the filtering problem. We deduce here a refined estimate of the rate of convergence of the approximation. We do this by proving a central limit type theorem for the error process. We also find the optimal variances of the Gaussian measures are of order $1/\sqrt{n}$. This implies, in particular, that the mean square error of the approximation as defined in [4, 5] is of order $1/n$.

math.PR↗

arXiv2013

Generalised Particle Filters with Gaussian Mixtures

Stochastic filtering is defined as the estimation of a partially observed dynamical system. A massive scientific and computational effort is dedicated to the development of numerical methods for approximating the solution of the filtering problem. Approximating the solution of the filtering problem with Gaussian mixtures has been a very popular method since the 1970s (see [1],[2],[46],[49]). Despite nearly fifty years of development, the existing work is based on the success of the numerical implementation and is not theoretically justified. This paper fills this gap and contains a rigorous analysis of a new Gaussian mixture approximation to the solution of the filtering problem. We deduce the L^2-convergence rate for the approximating system and show some numerical example to test the new algorithm.

math.PR↗

arXiv2013

Multiple Side-Band Generation for Two-Frequency Components Injected into a Tapered Amplifier

We have experimentally studied the multiple side-band generation for two-frequency components injected into a tapered amplifier and demonstrated its effects on atomic laser cooling. A heterodyne frequency-beat measurement and a Fabry Perot interferometer have been applied to analyze the side-band generation with different experimental parameters, such as frequency difference, injection laser power and tapered amplifier current. In laser cooling potassium40 and potassium41 with hyperfine splitting of 1.3GHz and 254MHz, respectively, the side-band generation with a small frequency difference has a significant effect on the number of trapped atoms.

physics.atom-ph↗

arXiv2013

Two particular EA-type binaries in the globular cluster omega Centauri

We analyzed the B and V light curves of two EA-type binaries V211 and NV358 using the WD code for the first time. Our analysis shows that V211 is a typical Algol-type binary and NV358 is a well detached binary system. As the two binaries are definite proper motion members of omega Centauri, we estimated their physical parameters, obtaining M_1=1.13\pm0.03M_\odot, R_1=0.98\pm0.01R_\odot and M_2=0.33\pm0.01M_\odot, R_2=0.92\pm0.01R_\odot for V211; M_1=1.30\pm0.05M_\odot, R_1=1.03\pm0.01R_\odot and M_2=0.58\pm0.02M_\odot, R_2=0.78\pm0.01R_\odot for NV358. On the color-magnitude diagram of omega Centauri, V211 is located in the faint blue straggler region and its primary component is more massive than a star at the main-sequence turnoff. Therefore, V211 is a blue straggler and should be formed by mass transfer from the secondary component to the primary. The age of NV358 is less than 1.93 Gyr, indicating that it is much younger than first-generation stars in omega Centauri. As NV364 in omega Centauri, NV358 should be a second-generation binary.

astro-ph.SR↗

arXiv2012

DQSB: A Reliable Broadcast Protocol Based on Distributed Quasi-Synchronized Mechanism for Low Duty-Cycled Wireless Sensor Networks

In duty-cycled wireless sensor networks, deployed sensor nodes are usually put to sleep for energy efficiency according to sleep scheduling approaches. Any sleep scheduling scheme with its supporting protocols ensures that data can always be routed from source to sink. In this paper, we investigate a problem of multi-hop broadcast and routing in random sleep scheduling scheme, and propose a novel protocol, called DQSB, by quasi-synchronization mechanism to achieve reliable broadcast and less latency routing. DQSB neither assumes time synchronization which requires all neighboring nodes wake up at the same time, nor assumes duty-cycled awareness which makes it difficult to use in asynchronous WSNs. Furthermore, the benefit of quasi-synchronized mechanism for broadcast from sink to other nodes is the less latency routing paths for reverse data collection to sink because of no or less sleep waiting time. Simulation results show that DQSB outperforms the existing protocols in broadcast times performance and keeps relative tolerant broadcast latency performance, even in the case of unreliable links. The proposed DQSB protocol, in this paper, can be recognized as a tradeoff between broadcast times and broadcast latency. We also explore the impact of parameters in the assumption and the approach to get proper values for supporting DQSB.

cs.NI↗

arXiv2012

Nonlinear PT-symmetric plaquettes

We introduce four basic two-dimensional (2D) plaquette configurations with onsite cubic nonlinearities, which may be used as building blocks for 2D PT -symmetric lattices. For each configuration, we develop a dynamical model and examine its PT symmetry. The corresponding nonlinear modes are analyzed starting from the Hamiltonian limit, with zero value of the gain-loss coefficient. Once the relevant waveforms have been identified (chiefly, in an analytical form), their stability is examined by means of linearization in the vicinity of stationary points. This reveals diverse and, occasionally, fairly complex bifurcations. The evolution of unstable modes is explored by means of direct simulations. In particular, stable localized modes are found in these systems, although the majority of identified solutions is unstable.

quant-ph↗

arXiv2011

Transfer and scattering of wave packets by a nonlinear trap

In the framework of a one-dimensional model with a tightly localized self-attractive nonlinearity, we study the formation and transfer (dragging) of a trapped mode by "nonlinear tweezers", as well as the scattering of coherent linear wave packets on the stationary localized nonlinearity. The use of the nonlinear trap for the dragging allows one to pick up and transfer the relevant structures without grabbing surrounding "garbage". A stability border for the dragged modes is identified by means of of analytical estimates and systematic simulations. In the framework of the scattering problem, the shares of trapped, reflected, and transmitted wave fields are found. Quasi-Airy stationary modes with a divergent norm, that may be dragged by the nonlinear trap moving at a constant acceleration, are briefly considered too.

nlin.PS↗