arXiv subjects

Nan Wang

Publications and source records attributed to Nan Wang.

210 records · Page 12Linked to original sources

arXiv2016

Challenges and Opportunities in Edge Computing

Many cloud-based applications employ a data centre as a central server to process data that is generated by edge devices, such as smartphones, tablets and wearables. This model places ever increasing demands on communication and computational infrastructure with inevitable adverse effect on Quality-of-Service and Experience. The concept of Edge Computing is predicated on moving some of this computational load towards the edge of the network to harness computational capabilities that are currently untapped in edge nodes, such as base stations, routers and switches. This position paper considers the challenges and opportunities that arise out of this new direction in the computing landscape.

cs.DC↗

arXiv2016

A Machine Learning Analysis of Twitter Sentiment to the Sandy Hook Shootings

Gun related violence is a complex issue and accounts for a large proportion of violent incidents. In the research reported in this paper, we set out to investigate the pro-gun and anti-gun sentiments expressed on a social media platform, namely Twitter, in response to the 2012 Sandy Hook Elementary School shooting in Connecticut, USA. Machine learning techniques are applied to classify a data corpus of over 700,000 tweets. The sentiments are captured using a public sentiment score that considers the volume of tweets as well as population. A web-based interactive tool is developed to visualise the sentiments and is available at http://www.gunsontwitter.com. The key findings from this research are: (i) There are elevated rates of both pro-gun and anti-gun sentiments on the day of the shooting. Surprisingly, the pro-gun sentiment remains high for a number of days following the event but the anti-gun sentiment quickly falls to pre-event levels. (ii) There is a different public response from each state, with the highest pro-gun sentiment not coming from those with highest gun ownership levels but rather from California, Texas and New York.

cs.SI↗

arXiv2016

Analytical formulation of lunar cratering asymmetries

We formulate the lunar cratering distribution and verify the cratering asymmetries generated by the main-belt asteroids (MBAs) as well as the near-Earth objects (NEOs). Based on a planar model that excludes the terrestrial and lunar gravitations on the impactors and assuming the impactor encounter speed with Earth $v_{\rm{enc}}$ is higher than the lunar orbital speed $v_{\rm{M}}$, we rigorously integrated the lunar cratering distribution, and derived its approximation to the first order of $v_{\rm{M}}/v_{\rm{enc}}$. Numerical simulations of lunar bombardment by the MBAs during the late heavy bombardment were performed with an Earth-Moon distance $a_{\rm{M}}$ = 20--60 Earth radii in five cases. The analytical model directly proves the existence of a leading/trailing asymmetry and the absence of near/far asymmetry. The approximate form of the leading/trailing asymmetry is $(1 + A_1 \cosβ)$, which decreases as the apex distance $β$ increases. The numerical simulations show evidence of a pole/equator asymmetry as well as the leading/trailing asymmetry, and the former is empirically described as $(1 + A_2 \cos2φ)$, which decreases as the latitude modulus $|φ|$ increases. The amplitudes $A_{1,2}$ are reliable measurements of asymmetries. Our analysis explicitly indicates the quantitative relations between cratering distribution and bombardment conditions (impactor properties and the lunar orbital status) like $A_1 \propto v_{\rm{M}}/v_{\rm{enc}}$, resulting in a method for reproducing the bombardment conditions through measuring the asymmetry. Mutual confirmation between analytical model and numerical simulations is found in terms of the cratering distribution and its variation with $a_{\rm{M}}$. Estimates of $A_1$ for crater density distributions generated by the MBAs and the NEOs are 0.101--0.159 and 0.117, respectively.

astro-ph.EP↗

arXiv2016

Determining proportions of lunar crater populations by fitting crater size distribution

We determine the proportions of two mixed crater populations distinguishable by size distributions on the Moon. A "multiple power-law" model is built to formulate crater size distribution $N(D) \propto D^{-α}$ whose slope $α$ varies with crater diameter $D$. Fitted size distribution of lunar highland craters characterized by $α= 1.17 \pm 0.04$, $1.88 \pm 0.07$, $3.17 \pm 0.10$ and $1.40 \pm 0.15$ for consecutive $D$ intervals divided by 49, 120 and 251 km and that of lunar Class 1 craters with a single slope $α= 1.96 \pm 0.14$, are taken as Population 1 and 2 crater size distribution respectively, whose sum is then fitted to the size distribution of global lunar craters with $D$ between 10 and 100 km. Estimated crater densities of Population 1 and 2 are $44 \times 10^{-5}$ and $5 \times 10^{-5}$ km$^{-2}$ respectively, leading to the proportion of the latter $10 \%$. The results underlines the need for considering the Population 1 craters and the relevant impactors, the primordial main-belt asteroids which dominated the late heavy bombardment.

astro-ph.EP↗

arXiv2015

A new boosting algorithm based on dual averaging scheme

The fields of machine learning and mathematical optimization increasingly intertwined. The special topic on supervised learning and convex optimization examines this interplay. The training part of most supervised learning algorithms can usually be reduced to an optimization problem that minimizes a loss between model predictions and training data. While most optimization techniques focus on accuracy and speed of convergence, the qualities of good optimization algorithm from the machine learning perspective can be quite different since machine learning is more than fitting the data. Better optimization algorithms that minimize the training loss can possibly give very poor generalization performance. In this paper, we examine a particular kind of machine learning algorithm, boosting, whose training process can be viewed as functional coordinate descent on the exponential loss. We study the relation between optimization techniques and machine learning by implementing a new boosting algorithm. DABoost, based on dual-averaging scheme and study its generalization performance. We show that DABoost, although slower in reducing the training error, in general enjoys a better generalization error than AdaBoost.

cs.LG↗

arXiv2014

Macroscopic models of collective motion with repulsion

We study a system of self-propelled particles which interact with their neighbors via alignment and repulsion. The particle velocities result from self-propulsion and repulsion by close neighbors. The direction of self-propulsion is continuously aligned to that of the neighbors, up to some noise. A continuum model is derived starting from a mean-field kinetic description of the particle system. It leads to a set of non conservative hydrodynamic equations. We provide a numerical validation of the continuum model by comparison with the particle model. We also provide comparisons with other self-propelled particle models with alignment and repulsion.

math-ph↗

arXiv2014

Modeling correlations in spontaneous activity of visual cortex with centered Gaussian-binary deep Boltzmann machines

Spontaneous cortical activity -- the ongoing cortical activities in absence of intentional sensory input -- is considered to play a vital role in many aspects of both normal brain functions and mental dysfunctions. We present a centered Gaussian-binary Deep Boltzmann Machine (GDBM) for modeling the activity in early cortical visual areas and relate the random sampling in GDBMs to the spontaneous cortical activity. After training the proposed model on natural image patches, we show that the samples collected from the model's probability distribution encompass similar activity patterns as found in the spontaneous activity. Specifically, filters having the same orientation preference tend to be active together during random sampling. Our work demonstrates the centered GDBM is a meaningful model approach for basic receptive field properties and the emergence of spontaneous activity patterns in early cortical visual areas. Besides, we show empirically that centered GDBMs do not suffer from the difficulties during training as GDBMs do and can be properly trained without the layer-wise pretraining.

cs.NE↗

arXiv2014

Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image Statistics

We present a theoretical analysis of Gaussian-binary restricted Boltzmann machines (GRBMs) from the perspective of density models. The key aspect of this analysis is to show that GRBMs can be formulated as a constrained mixture of Gaussians, which gives a much better insight into the model's capabilities and limitations. We show that GRBMs are capable of learning meaningful features both in a two-dimensional blind source separation task and in modeling natural images. Further, we show that reported difficulties in training GRBMs are due to the failure of the training algorithm rather than the model itself. Based on our analysis we are able to propose several training recipes, which allowed successful and fast training in our experiments. Finally, we discuss the relationship of GRBMs to several modifications that have been proposed to improve the model.

cs.NE↗

arXiv2012

Theoretical study of the synthesis of superheavy nuclei with Z= 119 and 120 in heavy-ion reactions with trans-uranium targets

By using a newly developed di-nuclear system model with a dynamical potential energy surface---the DNS-DyPES model, hot fusion reactions for synthesizing superheavy nuclei (SHN) with the charge number Z = 112-120 are studied. The calculated evaporation residue cross sections are in good agreement with available data. In the reaction 50Ti+249Bk -> (299-x)119 + xn, the maximal evaporation residue (ER) cross section is found to be about 0.11 pb for the 4n-emission channel. For projectile-target combinations producing SHN with Z=120, the ER cross section increases with the mass asymmetry in the incident channel increasing. The maximal ER cross sections for 58Fe+244Pu and 54Cr + 248Cm are relatively small (less than 0.01 pb) and those for 50Ti+249Cf and 50Ti+251Cf are about 0.05 and 0.25 pb, respectively.

nucl-th↗

arXiv2011

Anomaly Sequences Detection from Logs Based on Compression

Mining information from logs is an old and still active research topic. In recent years, with the rapid emerging of cloud computing, log mining becomes increasingly important to industry. This paper focus on one major mission of log mining: anomaly detection, and proposes a novel method for mining abnormal sequences from large logs. Different from previous anomaly detection systems which based on statistics, probabilities and Markov assumption, our approach measures the strangeness of a sequence using compression. It first trains a grammar about normal behaviors using grammar-based compression, then measures the information quantities and densities of questionable sequences according to incrementation of grammar length. We have applied our approach on mining some real bugs from fine grained execution logs. We have also tested its ability on intrusion detection using some publicity available system call traces. The experiments show that our method successfully selects the strange sequences which related to bugs or attacking.

cs.LG↗

arXiv2005

Particle transfer and fusion cross-section for Super-heavy nuclei in dinuclear system

Within the dinuclear system (DNS) conception, instead of solving Fokker-Planck Equation (FPE) analytically, the Master equation is solved numerically to calculate the fusion probability of super-heavy nuclei, so that the harmonic oscillator approximation to the potential energy of the DNS is avoided. The relative motion concerning the energy, the angular momentum, and the fragment deformation relaxations is explicitly treated to couple with the diffusion process, so that the nucleon transition probabilities, which are derived microscopically, are time-dependent. Comparing with the analytical solution of FPE, our results preserve more dynamical effects. The calculated evaporation residue cross sections for one-neutron emission channel of Pb-based reactions are basically in agreement with the known experimental data within one order of magnitude.

nucl-th↗

arXiv2005

Deformation and orientation effects in the driving potential of the dinuclear model

A double-folding method is used to calculate the nuclear and Coulomb interaction between two deformed nuclei with arbitrary orientations. A simplified Skryme-type interaction is adopted. The contributions of nuclear interaction and Coulomb interaction due to the deformation and orientation of the nuclei are evaluated for the driving potential used in the description of heavy-ion fusion reaction. So far there is no satisfactory theory to describe the evolution of the dynamical nuclear deformation and orientations during the heavy-ion fusion process. Our results estimated the magnitude of above effects.

nucl-th↗