arXiv subjects

Bo Zhang

Publications and source records attributed to Bo Zhang.

At least 847 records · Page 47Linked to original sources

arXiv2014

Polarization-dependent exciton dynamics in tetracene single crystals

We conduct polarization-dependent ultrafast spectroscopy to study the dynamics of singlet fission in tetracene single crystals. The spectrotemporal species for singlet and triplet excitons in transient absorption spectra are found to be strongly dependent on probe polarization. By carefully analyzing the polarization dependence, the signals contributed by different transitions related to singlet excitons have been disentangled, which is further applied to construct the correlation between dynamics of singlet and triplet excitons. The anisotropy of exciton dynamics provides an alternative approach to tackle the long-standing challenge in understanding the mechanism of singlet fission in organic semiconductors.

cond-mat.mtrl-sci↗

arXiv2014

Is Germanium (Ge, Z=32) A Neutron-Capture Element?

Historically,Ge has been considered to be a neutron-capture element. In this case, the r-process abundance of Ge is derived by subtracting the s-process abundance from the total abundance in the Solar system. However, the Ge abundance of the metal-poor star HD 108317 is lower than that of the scaled residual r-process abundance in the Solar system, about 1.2 dex. In this paper, based on a comparison of the Ge abundances of metal-poor stars and stellar yields, we find that the Ge abundances are not the result of the primary-like yields in massive stars and come mainly from the r-process. Based on the observed abundances of metal-poor stars, we derived the Ge abundances of the weak r-process and main r-process. The contributed percentage of the neutron-capture process to Ge in the Solar system is about 59 per cent, which means that the contributed percentage of the Ge residual abundance in the Solar system is about 41 per cent. We find that the Ge residual abundance is produced as secondary-like yields in massive stars. This implies that the element Ge in the Solar system is not produced solely by the neutron-capture process.

astro-ph.SR↗

arXiv2014

Through Skull Fluorescence Imaging of the Brain in a New Near-Infrared Window

To date, brain imaging has largely relied on X-ray computed tomography and magnetic resonance angiography with limited spatial resolution and long scanning times. Fluorescence-based brain imaging in the visible and traditional near-infrared regions (400-900 nm) is an alternative but currently requires craniotomy, cranial windows and skull thinning techniques, and the penetration depth is limited to 1-2 mm due to light scattering. Here, we report through-scalp and through-skull fluorescence imaging of mouse cerebral vasculature without craniotomy utilizing the intrinsic photoluminescence of single-walled carbon nanotubes in the 1.3-1.4 micrometre near-infrared window. Reduced photon scattering in this spectral region allows fluorescence imaging reaching a depth of >2 mm in mouse brain with sub-10 micrometre resolution. An imaging rate of ~5.3 frames/s allows for dynamic recording of blood perfusion in the cerebral vessels with sufficient temporal resolution, providing real-time assessment of blood flow anomaly in a mouse middle cerebral artery occlusion stroke model.

physics.med-ph↗

arXiv2014

Estimating R-Process Yields from Abundances of the Metal-Poor Stars

The chemical abundances of metal-poor stars provide important clues to explore stellar formation history and set significant constraints on models of the r-process. In this work, we find that the abundance patterns of the light and iron group elements of the main r-process stars are very close to those of the weak r-process stars. Based on a detailed abundance comparison, we find that the weak r-process occurs in supernovae with a progenitor mass range of $\sim11-26M_{\odot}$. Using the SN yields given by Heger & Woosley and the abundances of the weak r-process stars, the weak r-process yields are derived. The SNe with a progenitor mass range of $15M_{\odot}<M<26M_{\odot}$ are the main sites of the weak r-process and their contributions are larger than 80%. Using the abundance ratios of the weak r-process and the main r-process in the solar system, the average yields of the main r-process are estimated. The observed correlations of the [neutron-capture/Eu] versus [Eu/Fe] can be explained by mixing of the two r-process abundances in various fractions.

astro-ph.SR↗

arXiv2014

Ultra-Fast Fluorescence Imaging in Vivo with Conjugated Polymer Fluorophores in the Second Near-Infrared Window

In vivo fluorescence imaging in the second near-infrared window (1.0-1.7 microns) can afford deep tissue penetration and high spatial resolution, owing to the reduced scattering of long-wavelength photons. Here, we synthesize a series of low-bandgap donor/acceptor copolymers with tunable emission wavelengths of 1050-1350 nm in this window. Non-covalent functionalization with phospholipid-polyethylene glycol results in water-soluble and biocompatible polymeric nanoparticles, allowing for live cell molecular imaging at > 1000 nm with polymer fluorophores for the first time. Importantly, the high quantum yield of the polymer allows for in vivo, deep-tissue and ultrafast imaging of mouse arterial blood flow with an unprecedented frame rate of > 25 frames per second. The high time resolution results in spatially and time resolved imaging of the blood flow pattern in cardiogram waveform over a single cardiac cycle (~ 200 ms) of a mouse, which has not been observed with fluorescence imaging in this window before.

physics.bio-ph↗

arXiv2014

Superlinear density dependence of singlet fission rate in tetracene films

We experimentally show that the rate of singlet fission in tetracene films has a superlinear dependence on the density of photo-excited singlet excitons with ultrafast transient absorption spectroscopy. The spectrotemporal features of singlet and triplet dynamics can be disentangled from experimental data with the algorithm of singular value decomposition. The correlation between their temporal dynamics indicates a nonlinear density dependence of fission rate, which leads to a conjecture of coherent singlet fission process arising from superradiant excitons in crystalline tetracene. This hypothesis might be able to resolve some long-standing controversies.

cond-mat.mtrl-sci↗

arXiv2014

Dropout Training for Support Vector Machines

Dropout and other feature noising schemes have shown promising results in controlling over-fitting by artificially corrupting the training data. Though extensive theoretical and empirical studies have been performed for generalized linear models, little work has been done for support vector machines (SVMs), one of the most successful approaches for supervised learning. This paper presents dropout training for linear SVMs. To deal with the intractable expectation of the non-smooth hinge loss under corrupting distributions, we develop an iteratively re-weighted least square (IRLS) algorithm by exploring data augmentation techniques. Our algorithm iteratively minimizes the expectation of a re-weighted least square problem, where the re-weights have closed-form solutions. The similar ideas are applied to develop a new IRLS algorithm for the expected logistic loss under corrupting distributions. Our algorithms offer insights on the connection and difference between the hinge loss and logistic loss in dropout training. Empirical results on several real datasets demonstrate the effectiveness of dropout training on significantly boosting the classification accuracy of linear SVMs.

cs.LG↗

arXiv2014

Gamma-Ray Burst Prompt Emission Light Curves and Power Density Spectra in the ICMART Model

In this paper, we simulate the prompt emission light curves of gamma-ray bursts (GRBs) within the framework of the Internal-Collision-induced MAgnetic Reconnection and Turbulence (ICMART) model. This model applies to GRBs with a moderately-high magnetization parameter $σ$ in the emission region. We show that this model can produce highly variable light curves with both fast and slow components. The rapid variability is caused by many locally Doppler-boosted mini-emitters due to turbulent magnetic reconnection in a moderately-high-$σ$ flow. The run-away growth and subsequent depletion of these mini-emitters as a function time define a broad slow component for each ICMART event. A GRB light curve is usually composed of multiple ICMART events that are fundamentally driven by the erratic GRB central engine activity. Allowing variations of the model parameters, one is able to reproduce a variety of light curves and the power density spectra as observed.

astro-ph.HE↗

arXiv2014

Zero Dimensional Polariton Laser in a Sub-Wavelength Grating Based Vertical Microcavity

Semiconductor exciton-polaritons in planar microcavities form coherent two-dimensional condensates in non-equilibrium. However, coupling of multiple lower-dimensional polariton quantum systems, critically needed for polaritonic quantum device applications and novel cavity-lattice physics, has been limited due to the conventional cavity structures. Here we demonstrate full confinement of the polaritons non-destructively using a hybrid cavity made of a single-layer sub-wavelength grating mirror and a distributed Bragg reflector. Single-mode polariton lasing was observed at a chosen polarization. Incorporation of a designable slab mirror into the conventional vertical cavity, when operating in the strong-coupling regime, enables confinement, control and coupling of polariton gasses in a scalable fashion. It may open a door to experimental implementation of polariton-based quantum photonic devices and coupled cavity quantum electrodynamics systems.

cond-mat.quant-gas↗

arXiv2013

Magnetization Characteristic of Ferromagnetic Thin Strip by Measuring Anisotropic Magnetoresistance and Ferromagnetic Resonance

The magnetization characteristic in a permalloy thin strip is investigated by electrically measuring the anisotropic magnetoresistance and ferromagnetic resonance in in-plane and out-of-plane configurations. Our results indicate that the magnetization vector can rotate in the film plane as well as out of the film plane by changing the intensity of external magnetic field of certain direction. The magnetization characteristic can be explained by considering demagnetization and magnetic anisotropy. Our method can be used to obtain the demagnetization factor, saturated magnetic moment and the magnetic anisotropy.

cond-mat.mtrl-sci↗

arXiv2013

Gibbs Max-margin Topic Models with Data Augmentation

Max-margin learning is a powerful approach to building classifiers and structured output predictors. Recent work on max-margin supervised topic models has successfully integrated it with Bayesian topic models to discover discriminative latent semantic structures and make accurate predictions for unseen testing data. However, the resulting learning problems are usually hard to solve because of the non-smoothness of the margin loss. Existing approaches to building max-margin supervised topic models rely on an iterative procedure to solve multiple latent SVM subproblems with additional mean-field assumptions on the desired posterior distributions. This paper presents an alternative approach by defining a new max-margin loss. Namely, we present Gibbs max-margin supervised topic models, a latent variable Gibbs classifier to discover hidden topic representations for various tasks, including classification, regression and multi-task learning. Gibbs max-margin supervised topic models minimize an expected margin loss, which is an upper bound of the existing margin loss derived from an expected prediction rule. By introducing augmented variables and integrating out the Dirichlet variables analytically by conjugacy, we develop simple Gibbs sampling algorithms with no restricting assumptions and no need to solve SVM subproblems. Furthermore, each step of the "augment-and-collapse" Gibbs sampling algorithms has an analytical conditional distribution, from which samples can be easily drawn. Experimental results demonstrate significant improvements on time efficiency. The classification performance is also significantly improved over competitors on binary, multi-class and multi-label classification tasks.

stat.ML↗

arXiv2013

Improved Bayesian Logistic Supervised Topic Models with Data Augmentation

Supervised topic models with a logistic likelihood have two issues that potentially limit their practical use: 1) response variables are usually over-weighted by document word counts; and 2) existing variational inference methods make strict mean-field assumptions. We address these issues by: 1) introducing a regularization constant to better balance the two parts based on an optimization formulation of Bayesian inference; and 2) developing a simple Gibbs sampling algorithm by introducing auxiliary Polya-Gamma variables and collapsing out Dirichlet variables. Our augment-and-collapse sampling algorithm has analytical forms of each conditional distribution without making any restricting assumptions and can be easily parallelized. Empirical results demonstrate significant improvements on prediction performance and time efficiency.

cs.LG↗

arXiv2013

Discriminative Relational Topic Models

Many scientific and engineering fields involve analyzing network data. For document networks, relational topic models (RTMs) provide a probabilistic generative process to describe both the link structure and document contents, and they have shown promise on predicting network structures and discovering latent topic representations. However, existing RTMs have limitations in both the restricted model expressiveness and incapability of dealing with imbalanced network data. To expand the scope and improve the inference accuracy of RTMs, this paper presents three extensions: 1) unlike the common link likelihood with a diagonal weight matrix that allows the-same-topic interactions only, we generalize it to use a full weight matrix that captures all pairwise topic interactions and is applicable to asymmetric networks; 2) instead of doing standard Bayesian inference, we perform regularized Bayesian inference (RegBayes) with a regularization parameter to deal with the imbalanced link structure issue in common real networks and improve the discriminative ability of learned latent representations; and 3) instead of doing variational approximation with strict mean-field assumptions, we present collapsed Gibbs sampling algorithms for the generalized relational topic models by exploring data augmentation without making restricting assumptions. Under the generic RegBayes framework, we carefully investigate two popular discriminative loss functions, namely, the logistic log-loss and the max-margin hinge loss. Experimental results on several real network datasets demonstrate the significance of these extensions on improving the prediction performance, and the time efficiency can be dramatically improved with a simple fast approximation method.

cs.LG↗

arXiv2013

Investigation of the Puzzling Abundance Pattern in the Stars of the Fornax Dwarf Spheroidal Galaxy

Many works have found unusual characteristics of elemental abundances in nearby dwarf galaxies. This implies that there is a key factor of galactic evolution that is different from that of the Milky Way (MW). The chemical abundances of the stars in the Fornax dwarf spheroidal galaxy (Fornax dSph) provide excellent information for setting constraints on the models of the galactic chemical evolution. In this work, adopting the five-component approach, we fit the abundances of the Fornax dSph stars, including $α$ elements, iron group elements and neutron-capture elements. For most sample stars, the relative contributions from the various processes to the elemental abundances are not usually in the MW proportions. We find that the contributions from massive stars to the primary $α$ elements and iron group elements increase monotonously with increasing [Fe/H]. This means that the effect of the galactic wind is not strong enough to halt star formation and the contributions from massive stars to $α$ elements did not halted for [Fe/H]$\lesssim$-0.5. The average contributed ratios of various processes between the dSph stars and the MW stars monotonously decrease with increasing progenitor mass. This is important evidence of a bottom-heavy initial mass function (IMF) for the Fonax dSph, compared to the MW. Considering a bottom-heavy IMF for the dSph, the observed relations of [$α$/Fe] versus [Fe/H], [iron group/Fe] versus [Fe/H] and [neutron-capture/Fe] versus [Fe/H] for the dSph stars can be explained.

astro-ph.GA↗

arXiv2013

Lattice Boltzmann based discrete simulation for gas-solid fluidization

Discrete particle simulation, a combined approach of computational fluid dynamics and discrete methods such as DEM (Discrete Element Method), DSMC (Direct Simulation Monte Carlo), SPH (Smoothed Particle Hydrodynamics), PIC (Particle-In-Cell), etc., is becoming a practical tool for exploring lab-scale gas-solid systems owing to the fast development of parallel computation. However, gas-solid coupling and the corresponding fluid flow solver remain immature. In this work, we propose a modified lattice Boltzmann approach to consider the effect of both the local solid volume fraction and the local relative velocity between particles and fluid, which is different from the traditional volume-averaged Navier-Stokes equations. A time-driven hard sphere algorithm is combined to simulate the motion of individual particles, in which particles interact with each other via hard-sphere collisions, the collision detection and motion of particles are performed at constant time intervals. The EMMS (energy minimization multi-scale) drag is coupled with the lattice Boltzmann based discrete particle simulation to improve the accuracy. Two typical fluidization processes, namely, a single bubble injection at incipient fluidization and particle clustering in a fast fluidized bed riser, are simulated with this approach, with the results showing a good agreement with published correlations and experimental data. The capability of the approach to capture more detailed and intrinsic characteristics of particle-fluid systems is demonstrated. The method can also be used straightforward with other solid phase solvers.

physics.flu-dyn↗

arXiv2013

The study of s-process nucleosynthesis based on barium stars, CEMP-s and CEMP-r/s stars

In order to get a broader view of the s-process nucleosynthesis we study the abundance distribution of heavy elements of 35 barium stars and 24 CEMP-stars, including nine CEMP-s stars and 15 CEMP-r/s stars. The similar distribution of [Pb/hs] between CEMP-s and CEMP-r/s stars indicate that the s-process material of both CEMP-s and CEMP-r/s stars should have a uniform origin, i.e. mass transfer from their predominant AGB companions. For the CEMP-r/s stars, we found that the r-process should provide similar proportional contributes to the second s-peak and the third s-peak elements, and also be responsible for the higher overabundance of heavy elements than those in CEMP-s stars. Which hints that the r-process origin of CEMP-r/s stars should be closely linked to the main r-process. The fact that some small $r$ values exist for both barium and CEMP-s stars, implies that the single exposure event of the s-process nucleosynthesis should be general in a wide metallicity range of our Galaxy. Based on the relation between $C_{r}$ and $C_{s}$, we suggest that the origin of r-elements for CEMP-r/s stars have more sources. A common scenario is that the formation of the binary system was triggered by only one or a few supernova. In addition, accretion-induced collapse(AIC) or SN 1.5 should be the supplementary scenario, especially for these whose pre-AGB companion with higher mass and smaller orbit radius, which support the higher values of both $C_{r}$ and $C_{s}$.

astro-ph.SR↗

arXiv2013

Inverse electromagnetic scattering problems by a doubly periodic structure

Consider the problem of scattering of electromagnetic waves by a doubly periodic structure. The medium above the structure is assumed to be inhomogeneous characterized completely by an index of refraction. Below the structure is a perfect conductor or an imperfect conductor partially coated with a dielectric. Having established the well-posedness of the direct problem by the variational approach, we prove the uniqueness of the inverse problem, that is, the unique determination of the doubly periodic grating with its physical property and the index of refraction from a knowledge of the scattered near field by a countably infinite number of incident quasi-periodic electromagnetic waves. A key ingredient in our proofs is a novel mixed reciprocity relation derived in this paper.

math.AP↗

arXiv2013

A novel integral equation for scattering by locally rough surfaces and application to the inverse problem

This paper is concerned with the direct and inverse acoustic or electromagnetic scattering problems by a locally perturbed, perfectly reflecting, infinite plane (which is called a locally rough surface in this paper). We propose a novel integral equation formulation for the direct scattering problem which is defined on a bounded curve (consisting of a bounded part of the infinite plane containing the local perturbation and the lower part of a circle) with two corners. This novel integral equation can be solved efficiently by using the Nystrom method with a graded mesh introduced previously by Kress and is capable of dealing with large wavenumber cases. For the inverse problem, we propose a Newton iteration method to reconstruct the local perturbation of the plane from multiple frequency far-field data, based on the novel integral equation formulation. Numerical examples are carried out to demonstrate that our reconstruction method is stable and accurate even for the case of multiple-scale profiles.

math.AP↗