Showing 151–179 of 179 results for author: Hua, X

Homocentric Hypersphere Feature Embedding for Person Re-identification

Authors: Wangmeng Xiang, Jianqiang Huang, Xianbiao Qi, Xiansheng Hua, Lei Zhang

Abstract: Person re-identification (Person ReID) is a challenging task due to the large variations in camera viewpoint, lighting, resolution, and human pose. Recently, with the advancement of deep learning technologies, the performance of Person ReID has been improved swiftly. Feature extraction and feature matching are two crucial components in the training and deployment stages of Person ReID. However, ma… ▽ More Person re-identification (Person ReID) is a challenging task due to the large variations in camera viewpoint, lighting, resolution, and human pose. Recently, with the advancement of deep learning technologies, the performance of Person ReID has been improved swiftly. Feature extraction and feature matching are two crucial components in the training and deployment stages of Person ReID. However, many existing Person ReID methods have measure inconsistency between the training stage and the deployment stage, and they couple magnitude and orientation information of feature vectors in feature representation. Meanwhile, traditional triplet loss methods focus on samples within a mini-batch and lack knowledge of global feature distribution. To address these issues, we propose a novel homocentric hypersphere embedding scheme to decouple magnitude and orientation information for both feature and weight vectors, and reformulate classification loss and triplet loss to their angular versions and combine them into an angular discriminative loss. We evaluate our proposed method extensively on the widely used Person ReID benchmarks, including Market1501, CUHK03 and DukeMTMC-ReID. Our method demonstrates leading performance on all datasets. △ Less

Submitted 30 April, 2018; v1 submitted 24 April, 2018; originally announced April 2018.

Comments: 11 pages, 7 figures

Video2Shop: Exact Matching Clothes in Videos to Online Shopping Images

Authors: Zhi-Qi Cheng, Xiao Wu, Yang Liu, Xian-Sheng Hua

Abstract: In recent years, both online retail and video hosting service are exponentially growing. In this paper, we explore a new cross-domain task, Video2Shop, targeting for matching clothes appeared in videos to the exact same items in online shops. A novel deep neural network, called AsymNet, is proposed to explore this problem. For the image side, well-established methods are used to detect and extract… ▽ More In recent years, both online retail and video hosting service are exponentially growing. In this paper, we explore a new cross-domain task, Video2Shop, targeting for matching clothes appeared in videos to the exact same items in online shops. A novel deep neural network, called AsymNet, is proposed to explore this problem. For the image side, well-established methods are used to detect and extract features for clothing patches with arbitrary sizes. For the video side, deep visual features are extracted from detected object regions in each frame, and further fed into a Long Short-Term Memory (LSTM) framework for sequence modeling, which captures the temporal dynamics in videos. To conduct exact matching between videos and online shopping images, LSTM hidden states, representing the video, and image features, which represent static object images, are jointly modeled under the similarity network with reconfigurable deep tree structure. Moreover, an approximate training method is proposed to achieve the efficiency when training. Extensive experiments conducted on a large cross-domain dataset have demonstrated the effectiveness and efficiency of the proposed AsymNet, which outperforms the state-of-the-art methods. △ Less

Submitted 4 December, 2018; v1 submitted 14 April, 2018; originally announced April 2018.

Comments: IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2017

Global versus Localized Generative Adversarial Nets

Authors: Guo-Jun Qi, Liheng Zhang, Hao Hu, Marzieh Edraki, Jingdong Wang, Xian-Sheng Hua

Abstract: In this paper, we present a novel localized Generative Adversarial Net (GAN) to learn on the manifold of real data. Compared with the classic GAN that {\em globally} parameterizes a manifold, the Localized GAN (LGAN) uses local coordinate charts to parameterize distinct local geometry of how data points can transform at different locations on the manifold. Specifically, around each point there exi… ▽ More In this paper, we present a novel localized Generative Adversarial Net (GAN) to learn on the manifold of real data. Compared with the classic GAN that {\em globally} parameterizes a manifold, the Localized GAN (LGAN) uses local coordinate charts to parameterize distinct local geometry of how data points can transform at different locations on the manifold. Specifically, around each point there exists a {\em local} generator that can produce data following diverse patterns of transformations on the manifold. The locality nature of LGAN enables local generators to adapt to and directly access the local geometry without need to invert the generator in a global GAN. Furthermore, it can prevent the manifold from being locally collapsed to a dimensionally deficient tangent subspace by imposing an orthonormality prior between tangents. This provides a geometric approach to alleviating mode collapse at least locally on the manifold by imposing independence between data transformations in different tangent directions. We will also demonstrate the LGAN can be applied to train a robust classifier that prefers locally consistent classification decisions on the manifold, and the resultant regularizer is closely related with the Laplace-Beltrami operator. Our experiments show that the proposed LGANs can not only produce diverse image transformations, but also deliver superior classification performances. △ Less

Submitted 27 March, 2018; v1 submitted 16 November, 2017; originally announced November 2017.

Journal ref: Proceedings of of IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2018), Salt Lake City, Utah, June 18th - June 22nd, 2018

A Pilot Study of Domain Adaptation Effect for Neural Abstractive Summarization

Authors: Xinyu Hua, Lu Wang

Abstract: We study the problem of domain adaptation for neural abstractive summarization. We make initial efforts in investigating what information can be transferred to a new domain. Experimental results on news stories and opinion articles indicate that neural summarization model benefits from pre-training based on extractive summaries. We also find that the combination of in-domain and out-of-domain setu… ▽ More We study the problem of domain adaptation for neural abstractive summarization. We make initial efforts in investigating what information can be transferred to a new domain. Experimental results on news stories and opinion articles indicate that neural summarization model benefits from pre-training based on extractive summaries. We also find that the combination of in-domain and out-of-domain setup yields better summaries when in-domain data is insufficient. Further analysis shows that, the model is capable to select salient content even trained on out-of-domain data, but requires in-domain data to capture the style for a target domain. △ Less

Submitted 21 July, 2017; originally announced July 2017.

Comments: This paper is accepted by EMNLP 2017 Workshop on New Frontiers in Summarization

Understanding and Detecting Supporting Arguments of Diverse Types

Authors: Xinyu Hua, Lu Wang

Abstract: We investigate the problem of sentence-level supporting argument detection from relevant documents for user-specified claims. A dataset containing claims and associated citation articles is collected from online debate website idebate.org. We then manually label sentence-level supporting arguments from the documents along with their types as study, factual, opinion, or reasoning. We further charac… ▽ More We investigate the problem of sentence-level supporting argument detection from relevant documents for user-specified claims. A dataset containing claims and associated citation articles is collected from online debate website idebate.org. We then manually label sentence-level supporting arguments from the documents along with their types as study, factual, opinion, or reasoning. We further characterize arguments of different types, and explore whether leveraging type information can facilitate the supporting arguments detection task. Experimental results show that LambdaMART (Burges, 2010) ranker that uses features informed by argument types yields better performance than the same ranker trained without type information. △ Less

Submitted 2 May, 2017; v1 submitted 28 April, 2017; originally announced May 2017.

Comments: This paper is accepted as a short paper in ACL 2017

Refining Image Categorization by Exploiting Web Images and General Corpus

Authors: Yazhou Yao, Jian Zhang, Fumin Shen, Xiansheng Hua, Wankou Yang, Zhenmin Tang

Abstract: Studies show that refining real-world categories into semantic subcategories contributes to better image modeling and classification. Previous image sub-categorization work relying on labeled images and WordNet's hierarchy is not only labor-intensive, but also restricted to classify images into NOUN subcategories. To tackle these problems, in this work, we exploit general corpus information to aut… ▽ More Studies show that refining real-world categories into semantic subcategories contributes to better image modeling and classification. Previous image sub-categorization work relying on labeled images and WordNet's hierarchy is not only labor-intensive, but also restricted to classify images into NOUN subcategories. To tackle these problems, in this work, we exploit general corpus information to automatically select and subsequently classify web images into semantic rich (sub-)categories. The following two major challenges are well studied: 1) noise in the labels of subcategories derived from the general corpus; 2) noise in the labels of images retrieved from the web. Specifically, we first obtain the semantic refinement subcategories from the text perspective and remove the noise by the relevance-based approach. To suppress the search error induced noisy images, we then formulate image selection and classifier learning as a multi-class multi-instance learning problem and propose to solve the employed problem by the cutting-plane algorithm. The experiments show significant performance gains by using the generated data of our way on both image categorization and sub-categorization tasks. The proposed approach also consistently outperforms existing weakly supervised and web-supervised approaches. △ Less

Submitted 15 March, 2017; originally announced March 2017.

Exploiting Web Images for Dataset Construction: A Domain Robust Approach

Authors: Yazhou Yao, Jian Zhang, Fumin Shen, Xiansheng Hua, Jingsong Xu, Zhenmin Tang

Abstract: Labelled image datasets have played a critical role in high-level image understanding. However, the process of manual labelling is both time-consuming and labor intensive. To reduce the cost of manual labelling, there has been increased research interest in automatically constructing image datasets by exploiting web images. Datasets constructed by existing methods tend to have a weak domain adapta… ▽ More Labelled image datasets have played a critical role in high-level image understanding. However, the process of manual labelling is both time-consuming and labor intensive. To reduce the cost of manual labelling, there has been increased research interest in automatically constructing image datasets by exploiting web images. Datasets constructed by existing methods tend to have a weak domain adaptation ability, which is known as the "dataset bias problem". To address this issue, we present a novel image dataset construction framework that can be generalized well to unseen target domains. Specifically, the given queries are first expanded by searching the Google Books Ngrams Corpus to obtain a rich semantic description, from which the visually non-salient and less relevant expansions are filtered out. By treating each selected expansion as a "bag" and the retrieved images as "instances", image selection can be formulated as a multi-instance learning problem with constrained positive bags. We propose to solve the employed problems by the cutting-plane and concave-convex procedure (CCCP) algorithm. By using this approach, images from different distributions can be kept while noisy images are filtered out. To verify the effectiveness of our proposed approach, we build an image dataset with 20 categories. Extensive experiments on image classification, cross-dataset generalization, diversity comparison and object detection demonstrate the domain robustness of our dataset. △ Less

Submitted 28 March, 2017; v1 submitted 22 November, 2016; originally announced November 2016.

Comments: Journal

Deep CTR Prediction in Display Advertising

Authors: Junxuan Chen, Baigui Sun, Hao Li, Hongtao Lu, Xian-Sheng Hua

Abstract: Click through rate (CTR) prediction of image ads is the core task of online display advertising systems, and logistic regression (LR) has been frequently applied as the prediction model. However, LR model lacks the ability of extracting complex and intrinsic nonlinear features from handcrafted high-dimensional image features, which limits its effectiveness. To solve this issue, in this paper, we i… ▽ More Click through rate (CTR) prediction of image ads is the core task of online display advertising systems, and logistic regression (LR) has been frequently applied as the prediction model. However, LR model lacks the ability of extracting complex and intrinsic nonlinear features from handcrafted high-dimensional image features, which limits its effectiveness. To solve this issue, in this paper, we introduce a novel deep neural network (DNN) based model that directly predicts the CTR of an image ad based on raw image pixels and other basic features in one step. The DNN model employs convolution layers to automatically extract representative visual features from images, and nonlinear CTR features are then learned from visual features and other contextual features by using fully-connected layers. Empirical evaluations on a real world dataset with over 50 million records demonstrate the effectiveness and efficiency of this method. △ Less

Submitted 20 September, 2016; originally announced September 2016.

Comments: This manuscript is the accepted version for ACM Multimedia Conference 2016

The Spectral Energy Distributions of Fermi Blazars

Authors: J. H. Fan, J. H. Yang, Y. Liu, G. Y. Luo, C. Lin, Y. H. Yuan, H. B. Xiao, A. Y. Zhou, T. X. Hua, Z. Y. Pei

Abstract: (Abridged) In this paper, multi-wavelength data are compiled for a sample of 1425 Fermi blazars to calculate their spectral energy distributions (SEDs). A parabolic function, $\log(νF_ν) = P_1(\logν- P_2)^2 + P_3,$ is used for SED fitting. Synchrotron peak frequency ($\log ν_p$), spectral curvature ($P_1$), peak flux ($ν_{\rm p}F_{\rm ν_p}$), and integrated flux ($νF_ν$) are successfully obtained… ▽ More (Abridged) In this paper, multi-wavelength data are compiled for a sample of 1425 Fermi blazars to calculate their spectral energy distributions (SEDs). A parabolic function, $\log(νF_ν) = P_1(\logν- P_2)^2 + P_3,$ is used for SED fitting. Synchrotron peak frequency ($\log ν_p$), spectral curvature ($P_1$), peak flux ($ν_{\rm p}F_{\rm ν_p}$), and integrated flux ($νF_ν$) are successfully obtained for 1392 blazars (461 flat spectrum radio quasars-FSRQs, 620 BL Lacs-BLs and 311 blazars of uncertain type-BCUs, 999 sources have known redshifts). Monochromatic luminosity at radio 1.4 GHz, optical R band, X-ray at 1 keV and $γ$-ray at 1 GeV, peak luminosity, integrated luminosity and effective spectral indexes of radio to optical ($α_{\rm RO}$), and optical to X-ray ($α_{\rm OX}$) are calculated. The "Bayesian classification" is employed to log$ν_{\rm p}$ in the rest frame for 999 blazars with available redshift and the results show that 3 components are enough to fit the $\logν_{\rm p}$ distribution, there is no ultra high peaked subclass. Based on the 3 components, the subclasses of blazars using the acronyms of Abdo et al. (2010a) are classified, and some mutual correlations are also studied. Conclusions are finally drawn as follows: (1) SEDs are successfully obtained for 1392 blazars. The fitted peak frequencies are compared with common sources from samples available (Sambruna et al. 1996, Nieppola et al. 2006, 2008, Abdo et al. 2010a). (2) Blazars are classified as low synchrotron peak sources (LSPs) if $\logν_{\rm p}$(Hz) $\leq 14.0$, intermediate synchrotron peak sources (ISPs) if $14.0 < \logν_{\rm p}$(Hz) $\leq 15.3$, and high synchrotron peak sources (HSPs) if $\logν_{\rm p}$(Hz) $> 15.3$. (3) $γ$-ray emissions are strongly correlated with radio emissions. (...) △ Less

Submitted 13 August, 2016; originally announced August 2016.

Comments: ApJS accepted

The effect of randomness for dependency map on the robustness of interdependent lattices

Authors: Jing Yuan, Lixiang Li, Haipeng Peng, Jürgen Kurths, Xiaojing Hua, Yixian Yang

Abstract: For interdependent networks with identity dependency map, percolation is exactly the same with that on a single network and follows a second-order phase transition, while for random dependency, percolation follows a first-order phase transition. In real networks, the dependency relations between networks are neither identical nor completely random. Thus in this paper, we study the influence of ran… ▽ More For interdependent networks with identity dependency map, percolation is exactly the same with that on a single network and follows a second-order phase transition, while for random dependency, percolation follows a first-order phase transition. In real networks, the dependency relations between networks are neither identical nor completely random. Thus in this paper, we study the influence of randomness for dependency maps on the robustness of interdependent lattice networks. We introduce approximate entropy($ApEn$) as the measure of randomness of the dependency maps. We find that there is critical $ApEn_c$ below which the percolation is continuous, but for larger $ApEn$, it is a first-order transition. With the increment of $ApEn$, the $p_c$ increases until $ApEn$ reaching ${ApEn}_c'$ and then remains almost constant. The time scale of the system shows rich properties as $ApEn$ increases. Our results uncover that randomness is one of the important factors that lead to cascading failures of spatially interdependent networks. △ Less

Submitted 16 September, 2015; originally announced September 2015.

Image Tag Refinement by Regularized Latent Dirichlet Allocation

Authors: Jingdong Wang, Jiazhen Zhou, Hao Xu, Tao Mei, Xian-Sheng Hua, Shipeng Li

Abstract: Tagging is nowadays the most prevalent and practical way to make images searchable. However, in reality many manually-assigned tags are irrelevant to image content and hence are not reliable for applications. A lot of recent efforts have been conducted to refine image tags. In this paper, we propose to do tag refinement from the angle of topic modeling and present a novel graphical model, regulari… ▽ More Tagging is nowadays the most prevalent and practical way to make images searchable. However, in reality many manually-assigned tags are irrelevant to image content and hence are not reliable for applications. A lot of recent efforts have been conducted to refine image tags. In this paper, we propose to do tag refinement from the angle of topic modeling and present a novel graphical model, regularized Latent Dirichlet Allocation (rLDA). In the proposed approach, tag similarity and tag relevance are jointly estimated in an iterative manner, so that they can benefit from each other, and the multi-wise relationships among tags are explored. Moreover, both the statistics of tags and visual affinities of images in the corpus are explored to help topic modeling. We also analyze the superiority of our approach from the deep structure perspective. The experiments on tag ranking and image retrieval demonstrate the advantages of the proposed method. △ Less

Submitted 29 July, 2013; originally announced July 2013.

Hybrid Affinity Propagation

Authors: Jingdong Wang, Hao Xu, Xian-Sheng Hua, Shipeng Li

Abstract: In this paper, we address a problem of managing tagged images with hybrid summarization. We formulate this problem as finding a few image exemplars to represent the image set semantically and visually, and solve it in a hybrid way by exploiting both visual and textual information associated with images. We propose a novel approach, called homogeneous and heterogeneous message propagation (… ▽ More In this paper, we address a problem of managing tagged images with hybrid summarization. We formulate this problem as finding a few image exemplars to represent the image set semantically and visually, and solve it in a hybrid way by exploiting both visual and textual information associated with images. We propose a novel approach, called homogeneous and heterogeneous message propagation ($\text{H}^\text{2}\text{MP}$). Similar to the affinity propagation (AP) approach, $\text{H}^\text{2}\text{MP}$ reduce the conventional \emph{vector} message propagation to \emph{scalar} message propagation to make the algorithm more efficient. Beyond AP that can only handle homogeneous data, $\text{H}^\text{2}\text{MP}$ generalizes it to exploit extra heterogeneous relations and the generalization is non-trivial as the reduction to scalar messages from vector messages is more challenging. The main advantages of our approach lie in 1) that $\text{H}^\text{2}\text{MP}$ exploits visual similarity and in addition the useful information from the associated tags, including the associations relation between images and tags and the relations within tags, and 2) that the summary is both visually and semantically satisfactory. In addition, our approach can also present a textual summary to a tagged image collection, which can be used to automatically generate a textual description. The experimental results demonstrate the effectiveness and efficiency of the roposed approach. △ Less

Submitted 30 July, 2013; originally announced July 2013.

Beaming Effect in Fermi Blazars

Authors: J. H. Fan, J. H. Yang, J. Y. Zhang, T. X. Hua, Y. Liu, Y. P. Qin, Y. Huang

Abstract: The γ-ray loud blazars (flat spectrum radio quasars--FSRQs and BL Lacertae objects-BLs) are very bright in the γ-ray bands, which is perhaps associated with a beaming effect. Therefore, one can expect that the γ-ray luminosity is correlated with the beaming factor. In this paper, we investigated the relation between the radio Doppler factors and the gamma-ray luminosities. Our analysis suggests th… ▽ More The γ-ray loud blazars (flat spectrum radio quasars--FSRQs and BL Lacertae objects-BLs) are very bright in the γ-ray bands, which is perhaps associated with a beaming effect. Therefore, one can expect that the γ-ray luminosity is correlated with the beaming factor. In this paper, we investigated the relation between the radio Doppler factors and the gamma-ray luminosities. Our analysis suggests that the γ-ray luminosity be strongly correlated with the factor of δ_R for the whole sample, FSRQs, and BLs. When the effect of a common redshift is excluded, the correlation still exists for the FSRQs sub-sample suggesting that the γ-rays are strongly beamed. However, the partial correlation analysis does not show a correlation for the small BL Lac sample. △ Less

Submitted 15 October, 2012; originally announced October 2012.

Comments: 12 pages, 4 figures, PASJ (accepted)

Identification of gene pathways implicated in Alzheimer's disease using longitudinal imaging phenotypes with sparse regression

Authors: Matt Silver, Eva Janousova, Xue Hua, Paul M. Thompson, Giovanni Montana

Abstract: We present a new method for the detection of gene pathways associated with a multivariate quantitative trait, and use it to identify causal pathways associated with an imaging endophenotype characteristic of longitudinal structural change in the brains of patients with Alzheimer's disease (AD). Our method, known as pathways sparse reduced-rank regression (PsRRR), uses group lasso penalised regress… ▽ More We present a new method for the detection of gene pathways associated with a multivariate quantitative trait, and use it to identify causal pathways associated with an imaging endophenotype characteristic of longitudinal structural change in the brains of patients with Alzheimer's disease (AD). Our method, known as pathways sparse reduced-rank regression (PsRRR), uses group lasso penalised regression to jointly model the effects of genome-wide single nucleotide polymorphisms (SNPs), grouped into functional pathways using prior knowledge of gene-gene interactions. Pathways are ranked in order of importance using a resampling strategy that exploits finite sample variability. Our application study uses whole genome scans and MR images from 464 subjects in the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. 66,182 SNPs are mapped to 185 gene pathways from the KEGG pathways database. Voxel-wise imaging signatures characteristic of AD are obtained by analysing 3D patterns of structural change at 6, 12 and 24 months relative to baseline. High-ranking, AD endophenotype-associated pathways in our study include those describing chemokine, Jak-stat and insulin signalling pathways, and tight junction interactions. All of these have been previously implicated in AD biology. In a secondary analysis, we investigate SNPs and genes that may be driving pathway selection, and identify a number of previously validated AD genes including CR1, APOE and TOMM40. △ Less

Submitted 9 April, 2012; originally announced April 2012.

A possible disk mechanism for the 23d QPO in Mkn~501

Authors: J. H. Fan, F. M. Rieger, T. X. Hua, U. C. Joshi, J. Li, Y. X. Wang, J. L. Zhou, Y. H. Yuan, J. B. Su, Y. W. Zhang

Abstract: Optically thin two-temperature accretion flows may be thermally and viscously stable, but acoustically unstable. Here we propose that the O-mode instability of a cooling-dominated optically thin two-temperature inner disk may explain the 23-day quasi-periodic oscillation (QPO) period observed in the TeV and X-ray light curves of Mkn~501 during its 1997 high state. In our model the relativistic j… ▽ More Optically thin two-temperature accretion flows may be thermally and viscously stable, but acoustically unstable. Here we propose that the O-mode instability of a cooling-dominated optically thin two-temperature inner disk may explain the 23-day quasi-periodic oscillation (QPO) period observed in the TeV and X-ray light curves of Mkn~501 during its 1997 high state. In our model the relativistic jet electrons Compton upscatter the disk soft X-ray photons to TeV energies, so that the instability-driven X-ray periodicity will lead to a corresponding quasi-periodicity in the TeV light curve and produce correlated variability. We analyse the dependence of the instability-driven quasi-periodicity on the mass (M) of the central black hole, the accretion rate ($\rm{\dot{M}}$) and the viscous parameter ($α$) of the inner disk. We show that in the case of Mkn~501 the first two parameters are constrained by various observational results, so that for the instability occurring within a two-temperature disk where $α=0.05-1.0$, the quasi-period is expected to lie within the range of 8 to 100 days, as indeed the case. In particular, for the observed 23-day QPO period our model implies a viscosity coefficient $α\leq 0.28$, a sub-Eddington accretion rate $\dot{M} \simeq 0.02 \dot{M}_{\rm Edd}$ and a transition radius to the outer standard disk of $r_0 \sim 60 r_g$, and predicts a period variation $δP/P \sim 0.23$ due to the motion of the instability region. △ Less

Submitted 5 October, 2007; originally announced October 2007.

Comments: 18 pages, 1 figure, accepted by APh

Journal ref: Astropart.Phys.28:508-515,2008

Derivations for the even part of the odd Hamiltonian superalgebra in positive characteristic

Authors: Xiuying Hua, Yucai Su

Abstract: In this paper we mainly study the derivations for even part of the finite-dimensional odd Hamiltonian superalgebra $HO$ over a field of prime characteristic. We first give the generator set of $HO_{\bar{0}}.$ Then we determine the homogeneous derivationsfrom $HO_{\bar{0}}$ into $W_{\bar{0}},$ the even part of the generalized Witt superalgebra $W$. Finally, wedetermine the derivation algebra and… ▽ More In this paper we mainly study the derivations for even part of the finite-dimensional odd Hamiltonian superalgebra $HO$ over a field of prime characteristic. We first give the generator set of $HO_{\bar{0}}.$ Then we determine the homogeneous derivationsfrom $HO_{\bar{0}}$ into $W_{\bar{0}},$ the even part of the generalized Witt superalgebra $W$. Finally, wedetermine the derivation algebra and outer derivation algebra of $HO_{\bar{0}}.$ △ Less

Submitted 26 October, 2005; v1 submitted 26 October, 2005; originally announced October 2005.

Comments: 30 pages

MSC Class: 17B50; 17B40

Journal ref: Acta Math Sin. 25 (2009) 355-378

Electron self-energy and effective mass in a single heterostructure

Authors: Xiu-Kun Hua, Yin-Zhong Wu, Zhen-Ya Li

Abstract: In this paper, we investigate the electron self-energy and effective mass in a single heterostructure using Green-function method. Numerical calculations of the electron self-energy and effective mass for GaAs/AlAs heterostructure are performed. The results show that the self energy (effective mass) of electron, which incorporate the energy of electron coupling to interface-optical phonons and h… ▽ More In this paper, we investigate the electron self-energy and effective mass in a single heterostructure using Green-function method. Numerical calculations of the electron self-energy and effective mass for GaAs/AlAs heterostructure are performed. The results show that the self energy (effective mass) of electron, which incorporate the energy of electron coupling to interface-optical phonons and half three-dimension LO phonons, monotonically increase(decrease) from that of interface polaron to that of 3D bulk polaron with the increase of the distance between the position of the electron and interface. △ Less

Submitted 20 May, 2002; originally announced May 2002.

Comments: 10 pages, 2 figures

Modeling The Time Variability of Accreting Compact Sources

Authors: Demosthenes Kazanas, Xin-Min Hua

Abstract: We present model light curves for accreting Black Hole Candidates (BHC) based on a recently proposed model for their spectro-temporal properties. According to this model, the observed light curves and aperiodic variability of BHC are due to a series of soft photon injections at random (Poisson) intervals near the compact object and their reprocessing into hard radiation in an extended but non-un… ▽ More We present model light curves for accreting Black Hole Candidates (BHC) based on a recently proposed model for their spectro-temporal properties. According to this model, the observed light curves and aperiodic variability of BHC are due to a series of soft photon injections at random (Poisson) intervals near the compact object and their reprocessing into hard radiation in an extended but non-uniform hot plasma corona surrounding the compact object. We argue that the majority of the timing characteristics of these light curves are due to the stochastic nature of the Comptonization process in the extended corona, whose properties, most notably its radial density dependence, are imprinted in them. We compute the corresponding Power Spectral Densities (PSD), autocorrelation functions, time skewness of the light curves and time lags between the light curves of the sources at different photon energies and compare our results to observation. Our model light curves compare well with observations, providing good fits to their overall morphology, as manifest by the autocorrelation and skewness functions. The lags and PSDs of the model light curves are also in good agreement with those observed (the model can even accommodate the presence of QPOs). Finally, while most of the variability power resides at time scales $\gsim$ a few seconds, at the same time, the model allows also for shots of a few msec in duration, in accordance with observation. We suggest that refinements of this type of model along with spectral and phase lag information can be used to probe the structure of this class of high energy sources. △ Less

Submitted 12 February, 1999; originally announced February 1999.

Comments: 23 pages Latex, 15 encapsulated postscript figures, to appear in the Astrophysical Journal

Probing the Structure of Accreting Compact Sources Through X-Ray Time Lags and Spectra

Authors: Xin-Min Hua, Demosthenes Kazanas, Wei Cui

Abstract: We exhibit, by compiling all data sets we can acquire, that the Fourier frequency dependent hard X-ray lags, first observed in the analysis of aperiodic variability of the light curves of the black hole candidate Cygnus X-1, appear to be a property shared by several other accreting black hole candidate sources and also by the different spectral states of this source. We then present both analyti… ▽ More We exhibit, by compiling all data sets we can acquire, that the Fourier frequency dependent hard X-ray lags, first observed in the analysis of aperiodic variability of the light curves of the black hole candidate Cygnus X-1, appear to be a property shared by several other accreting black hole candidate sources and also by the different spectral states of this source. We then present both analytic and numerical models of these time lags resulting by the process of Comptonization in a variety of hot electron configurations. We argue that under the assumption that the observed spectra are due to Comptonization, the dependence of the lags on the Fourier period provides a means for mapping the spatial density profile of the hot electron plasma, while the period at which the lags eventually level--off provides an estimate of the size of the scattering cloud. We further examine the influence of the location and spatial extent of the soft photon source on the form of the resulting lags for a variety of configurations; we conclude that the study of the X-ray hard lags can provide clues about these parameters of the Comptonization process too. Fits of the existing data with our models indicate that the size of the Comptonizing clouds are quite large in extent ($\sim$ 1 light second) with inferred radial density profiles which are in many instances inconsistent with those of the standard dynamical models, while the extent of the source of soft photons appears to be much smaller than those of the hot electrons by roughly two orders of magnitude and its location consistent with the center of the hot electron corona. △ Less

Submitted 8 October, 1998; originally announced October 1998.

Comments: 20 pages Latex, 11 postscript figures, to appear in the Astrophysical Journal, Vol 512, Feb 20 issue

Timing Noise Properties of GRO J0422+32

Authors: J. E. Grove, M. S. Strickman, S. M. Matz, X. -M. Hua, D. Kazanas, L. Titarchuk

Abstract: OSSE observed the transient black hole candidate GRO J0422+32 (XN Per 92) between 1992 Aug 11 and 1992 Sep 17. High time resolution data were obtained in several energy bands over the 35-600 keV range with a sampling rate of 8 ms. Power spectra at energies below 175 keV show substantial low-frequency red noise with a shoulder at a few tens of mHz, peaked noise with characteristic frequency near… ▽ More OSSE observed the transient black hole candidate GRO J0422+32 (XN Per 92) between 1992 Aug 11 and 1992 Sep 17. High time resolution data were obtained in several energy bands over the 35-600 keV range with a sampling rate of 8 ms. Power spectra at energies below 175 keV show substantial low-frequency red noise with a shoulder at a few tens of mHz, peaked noise with characteristic frequency near 0.2 Hz, and a second shoulder at a few Hz. The frequencies of the shoulders and the peak are independent of energy and source intensity. The complex cross spectrum indicates that photons in the 75-175 keV band lag photons in the 35-60 keV band by a time roughly proportional to the inverse of the Fourier frequency. The maximum lag observed is about 300 ms. The power and lag spectra are consistent with the production of the gamma rays through thermal Comptonization in an extended hot corona with a power-law density profile. △ Less

Submitted 19 May, 1998; originally announced May 1998.

Comments: 14 pages, 2 figs, uses psfig, accepted for pub in ApJ Letters

Galactic Black Hole Binaries: High-Energy Radiation

Authors: J. E. Grove, J. E. Grindlay, B. A. Harmon, X. -M. Hua, D. Kazanas, M. McConnell

Abstract: Observations of galactic BHCs made by the Compton GRO in the hard X-ray and gamma-ray bands have significantly enhanced our knowledge of the emission properties of these objects. Understanding these observations presents a formidable challenge to theoretical models of the accretion flow onto the compact object and of the physical mechanisms that generate high-energy radiation. Here we summarize… ▽ More Observations of galactic BHCs made by the Compton GRO in the hard X-ray and gamma-ray bands have significantly enhanced our knowledge of the emission properties of these objects. Understanding these observations presents a formidable challenge to theoretical models of the accretion flow onto the compact object and of the physical mechanisms that generate high-energy radiation. Here we summarize the current state of observations and theoretical interpretation of the emission from BHCs above 20 keV. The all-sky monitoring capability of BATSE allows nearly continuous studies of the high-energy emission from more than a dozen BHCs. These long-term datasets are particularly well-suited to multi-wavelength studies. Energy spectral evolution and/or state transitions have been observed from many of the BHCs. Observations above 50 keV from OSSE demonstrate the existence of two gamma ray spectral states that appear to be the extensions of the X-ray low/hard and high/soft (or perhaps very high) states. The former state, the "breaking" state, cuts off with e-folding energy ~100 keV and has its peak luminosity near this energy. In contrast, the latter state has luminosity peaking in the soft X-rays and an unbroken power law spectrum, even up to energies above 500 keV in some cases. The breaking gamma-ray spectrum can be well modeled by Comptonization of soft photons in a hot thermal plasma. The power-law state creates more significant theoretical challenges. It has been suggested that in this state the high- energy emission arises from bulk-motion Comptonization in the convergent accretion flow from the inner edge of the accretion disk. △ Less

Submitted 24 April, 1998; originally announced April 1998.

Comments: 19 pages, 6 figures, prepared with aipproc.sty

Journal ref: AIP Conf.Proc.410:122,1997

Frequency Dependent Lags - A Common Phenomenon of Accreting Sources

Authors: Xin-Min Hua, Demosthenes Kazanas, Wei Cui

Abstract: The Fourier frequency dependent hard X-ray lag, first discovered from the analysis of aperiodic variability of the light curves of the black hole candidate Cygnus X-1, turns out to be a property shared by several other accreting compact sources. We show that the lag can be explained in terms of Comptonization process in coronae of hot electrons with inhomogeneous density distributions. The densi… ▽ More The Fourier frequency dependent hard X-ray lag, first discovered from the analysis of aperiodic variability of the light curves of the black hole candidate Cygnus X-1, turns out to be a property shared by several other accreting compact sources. We show that the lag can be explained in terms of Comptonization process in coronae of hot electrons with inhomogeneous density distributions. The density profile of a corona, like the optical depth and electron temperature, can significantly affect the Comptonization energy spectrum from it. This means, by fitting the energy spectrum alone, it is not possible to uniquely determine the optical depth and temperature of the Comptonization cloud if its density profile is unknown. The hard X-ray time lag is sensitive to the density distribution of the scattering corona. Thus simultaneous analysis of the spectral and temporal X-ray data will allow us to probe the density structure of the Comptonizing corona and thereby the dynamics of mass accretion onto the compact object. △ Less

Submitted 17 December, 1997; originally announced December 1997.

Comments: 4 pages, 3 figures, Tex file, using AIP Proceedings macro, submitted to Proceedings of the 8th Annual October Astrophysics Conference in Maryland (Dec. 15, 1997)

Modeling the Time Variability of Black Hole Candidates

Authors: Demosthenes Kazanas, Xin-Min Hua

Abstract: We present model light curves, Power Spectral Densities (PSD) and time lags of accreting Black Hole Candidates (BHC) based on a recent model of these sources. According to our model the observed variability is due, to a large extent, to the stochastic nature of Comptonization, the basic process responsible also for the emission of high energy radiation. Our additional assumption is that the Comp… ▽ More We present model light curves, Power Spectral Densities (PSD) and time lags of accreting Black Hole Candidates (BHC) based on a recent model of these sources. According to our model the observed variability is due, to a large extent, to the stochastic nature of Comptonization, the basic process responsible also for the emission of high energy radiation. Our additional assumption is that the Comptonization process takes place in an extended but non-uniform atmosphere around the compact object. Our model reproduces the observed light curves well, in that it provides a good fit to the PSDs and the overall light curve morphology, indicating, in accordance with observation, that most of the power resides at time scales $\gsim$ a few seconds while at the same time one can distinguish shots of a few msec in duration. We suggest that refinement of this type of model along with spectral and phase lag information can be used to probe the structure of this class of high energy sources. △ Less

Submitted 15 December, 1997; originally announced December 1997.

Comments: Contributed paper to 8th Annual October Astrophysics Conference in Maryland, "Accretion Processes in Astrophysical Systems: Some Like It Hot", 4 pages latex, 2 .eps figures, aipproc.sty, available at ftp://ftp.aip.org/ems/tex/macros/proceedings/8.5x11

Cygnus X-1: X-ray Emission Mechanism and Geometry

Authors: Wei Cui, S. N. Zhang, J. H. Swank, X. -M. Hua, K. Ebisawa, T. Dotani

Abstract: We summarize recent observations of Cyg X-1 during its 1996 state transition. The results favor an emission geometry with a central optically thin corona surrounded by an optically thick, geometrically thin accretion disk. As the soft state is approached, the inner edge of the disk moves closer to the black hole, and the corona shrinks, perhaps due to more efficient cooling. Soft X-rays are from… ▽ More We summarize recent observations of Cyg X-1 during its 1996 state transition. The results favor an emission geometry with a central optically thin corona surrounded by an optically thick, geometrically thin accretion disk. As the soft state is approached, the inner edge of the disk moves closer to the black hole, and the corona shrinks, perhaps due to more efficient cooling. Soft X-rays are from the innermost disk, and inverse-Comptionization by the corona appears to be responsible for the hard X-ray emission. Both Compton reflection and an iron emission line have been observed in both states. They are likely to originate in the inner region of the disk. Also briefly discussed is the theoretical effort to simultaneously model the observed spectral and temporal X-ray properties of Cyg X-1. △ Less

Submitted 14 November, 1997; originally announced November 1997.

Comments: 2 pages. Proceedings of IAU Symposium 188 ``The Hot Universe'', Kyoto, Japan, August 26-30, 1997

Modeling the Hard X-Ray Lag and Energy Spectrum of Cyg X-1

Authors: Xin-Min Hua, Demosthenes Kazanas, Wei Cui

Abstract: In an effort to model the observed energy spectrum of Cygnus X-1 as well as its hard X-ray lag by Comptonization in inhomogeneous clouds of hot electrons with spherical geometry and various radial density profiles we discovered that: 1) Plasma clouds with different density profiles will lead to different Comptonization energy spectra even though they have the same optical depth and temperature.… ▽ More In an effort to model the observed energy spectrum of Cygnus X-1 as well as its hard X-ray lag by Comptonization in inhomogeneous clouds of hot electrons with spherical geometry and various radial density profiles we discovered that: 1) Plasma clouds with different density profiles will lead to different Comptonization energy spectra even though they have the same optical depth and temperature. On the other hand, clouds with different optical depths can produce the same energy spectra as long as their radial density distributions are properly chosen. Thus by fitting the energy spectrum alone, it is not possible to uniquely determine the optical depth of the Comptonization cloud, let alone its density structure. 2) The phase or time difference as a function of Fourier frequency or period for the X-rays in two energy bands is sensitive to the radial density distribution of the scattering cloud. Comptonization in plasma clouds with non-uniform density profiles can account for the long standing puzzle of the frequency-dependent hard X-ray lags of Cygnus X-1 and other sources. Thus simultaneously fitting the observed spectral and temporal X-ray properties will allow us to probe the density structure of the Comptonizing atmosphere and thereby the dynamics of mass accretion onto the compact object. △ Less

Submitted 16 October, 1997; originally announced October 1997.

Comments: 8 pages, 5 figures, Tex file, using AAS macro, submitted to Astrophysical Journal (June 24, 1997)

Monte Carlo Simulation of Comptonization in Inhomogeneous Media

Authors: Xin-Min Hua

Abstract: Comptonization is the process in which photon spectrum changes due to multiple Compton scatterings in the electronic plasma. It plays an important role in the spectral formation of astrophysical X-ray and gamma-ray sources. There are several intrinsic limitations for the analytical method in dealing with the Comptonization problem and Monte Carlo simulation is one of the few alternatives. We des… ▽ More Comptonization is the process in which photon spectrum changes due to multiple Compton scatterings in the electronic plasma. It plays an important role in the spectral formation of astrophysical X-ray and gamma-ray sources. There are several intrinsic limitations for the analytical method in dealing with the Comptonization problem and Monte Carlo simulation is one of the few alternatives. We describe an efficient Monte Carlo method that can solve the Comptonization problem in a fully relativistic way. We expanded the method so that it is capable of simulating Comptonization in the media where electron density and temperature varies discontinuously from one region to the other and in the isothermal media where density varies continuously along photon paths. The algorithms are presented in detail to facilitate computer code implementation. We also present a few examples of its application to the astrophysical research. △ Less

Submitted 16 September, 1997; originally announced September 1997.

Comments: 12 pages, 4 figures, Postscript file, in press ("Computers in Physics", Vol. 11, No. 6)

The Duration -- Photon Energy Relation of Gamma Ray Bursts and Its Interpretations

Authors: Demosthenes Kazanas, Lev G. Titarchuk, Xin-Min Hua

Abstract: Fenimore et al. (1995) have recently presented a very tight correlation between the spectral and the temporal structure in Gamma Ray Bursts (hereafter GRBs). In particular, they discovered that the durations of the constituent subpulses which make up the time profile of a given GRB have a well defined power-law dependence, of index $\simeq 0.45$, on the energy $E$ of the observed photons. In thi… ▽ More Fenimore et al. (1995) have recently presented a very tight correlation between the spectral and the temporal structure in Gamma Ray Bursts (hereafter GRBs). In particular, they discovered that the durations of the constituent subpulses which make up the time profile of a given GRB have a well defined power-law dependence, of index $\simeq 0.45$, on the energy $E$ of the observed photons. In this note we present two simple models which can account in a straightforward fashion for the observed correlation. These models involve: (a) The impulsive injection of a population of relativistic electrons and their subsequent cooling by synchrotron radiation. (b) The impulsive injection of monoenergetic high energy photons in a medium of Thompson depth $τ_T \sim 5$ and their subsequent downgrading in energy due to electron scattering. We present arguments for distinguishing between these two models from the existing data. △ Less

Submitted 18 September, 1997; originally announced September 1997.

Comments: 13 pages TeX, Accepted for Publication in the Astrophysical Journal

Phase Difference and Coherence as Diagnistics of Accreting Compact Sources

Authors: Xin-Min Hua, Demosthenes Kazanas, Lev Titarchuk

Abstract: We present calculations of the time lags and the coherence function of X-ray photons for a novel model of radiation emission from accretion powered, high-energy sources. Our model involves only Comptonization of soft photons injected near the compact object in an extended but non-uniform atmosphere around the compact object. Our results show that this model produces time lags between the hard an… ▽ More We present calculations of the time lags and the coherence function of X-ray photons for a novel model of radiation emission from accretion powered, high-energy sources. Our model involves only Comptonization of soft photons injected near the compact object in an extended but non-uniform atmosphere around the compact object. Our results show that this model produces time lags between the hard and soft bands of the X-ray spectrum which increase with Fourier period, in agreement with recent observations; it also produces a coherence function equal to one over a wide range of frequencies if the system parameters do not have significant changes, also in agreement with the limited existing observations. We explore various conditions that could affect coherence functions. We indicate that measurements of these statistical quantities could provide diagnostics of the radial structure of the density of this class of sources. △ Less

Submitted 18 March, 1997; originally announced March 1997.

Comments: 15 pages, Latex, using AAS macro, to be published in Astrophysical Journal Letters.

Temporal and Spectral Properties of Comptonized Radiation

Authors: Demos Kazanas, Xin-Min Hua, Lev Titarchuk

Abstract: We have found relations between the temporal and spectral properties of radiation Comptonized in an extended atmosphere associated with compact accreting sources. We demonstrate that the fluctuation power spectrum density (PSD) imposes constraints on the atmosphere scale and profile. Furthermore, we indicate that the slope and low frequency break of the PSD are related to the Thomson depth tau_0… ▽ More We have found relations between the temporal and spectral properties of radiation Comptonized in an extended atmosphere associated with compact accreting sources. We demonstrate that the fluctuation power spectrum density (PSD) imposes constraints on the atmosphere scale and profile. Furthermore, we indicate that the slope and low frequency break of the PSD are related to the Thomson depth tau_0 of the atmosphere and the radius of its physical size respectively. Since the energy spectrum of the escaping radiation depends also on tau_0 (and the electron temperature kT_e), the relation between spectral and temporal properties follows. This relation allows for the first time an estimate of the accreting matter Thomson depth tau_0 independent of arguments involving Comptonization. We present figures for the light curves and PSD of different energy bands, the photon energy spectra and the phase lags as functions of the variability frequency. The temporal properties of the high (soft) and low (hard) state of black hole sources are discussed in this context. △ Less

Submitted 19 December, 1996; originally announced December 1996.

Comments: 18 pages, Latex, using AAS macro, to be published in Astrophysical Journal. A condensed version to appear in the Proceedings of Integral meeting.