检索结果-内蒙古大学图书馆

Heterogeneous memory enhanced graph reasoning network for cross-modal retrieval

Science China(Information Sciences) 2022年第7期65卷 157-169页

作者： Zhong JI Kexin CHEN Yuqing HE Yanwei PANG Xuelong LI School of Electrical and Information Engineering Tianjin University Center for Optical Imagery Analysis and Learning Northwestern Polytechnical University

Cross-modal retrieval(CMR) aims to retrieve the instances of a specific modality that are relevant to a given query from another modality, which has drawn much attention because of its importance in bridging vision with language. A key to the success of CMR is to learn more discriminative and robust representations for both visual and textual instances to further reduce the heterogeneous discrepancy existing in different modalities. In this paper, we address this challenging issue by proposing a heterogeneous memory enhanced graph reasoning network, named HMGR, to connect the semantic correlations between vision and *** the one hand, we design a novel dual-path network architecture to generate relationship enhanced global representations by employing modality-specific graph reasoning on extracted local features for each *** this way, the topological interdependencies of both visual and textual intra-instance local fragments are fully mined to achieve a deeper semantic understanding of the relationships between them. On the other hand,we focus on utilizing inter-instance semantic correlated knowledge to enhance the discriminability of the final learned representations, which is achieved by introducing a joint heterogeneous memory network to iteratively restore both visual and textual instance-level information. Through interacting with long-term contextual multimodal knowledge, an encouraging shared latent feature space for mitigating the heterogeneous gap across different modalities can be learned. Extensive experiments under both image-text retrieval and video-text retrieval scenarios on three benchmark datasets demonstrate the effectiveness of our proposed method.

关键词： cross-modal retrieval graph reasoning memory network visual semantic embedding imagetext retrieval video-text retrieval

来源：评论

学校读者我要写书评

暂无评论

Direct measurement of dipoles in anomalous elasticity of amorphous solids

引用

Physical Review E 2022年第4期105卷 L043001-L043001页

作者： Bhanu Prasad Bhowmik Michael Moshe Itamar Procaccia Department of Chemical Physics The Weizmann Institute of Science Rehovot 76100 Israel Racah Institute of Physics The Hebrew University of Jerusalem Jerusalem 9190 Israel Center for Optical Imagery Analysis and Learning Northwestern Polytechnical University Xi'an 710072 China

Recent progress in studying the physics of amorphous solids has revealed that mechanical strains can be strongly screened by the formation of plastic events that are typically quadrupolar in nature. The theory stipulates that gradients in the density of the quadrupoles act as emergent dipole sources, leading to strong screening and to qualitative changes in the mechanical response, as seen, for example, in the displacement field. In this Letter we first offer direct measurements of the dipole field, independently of any theoretical assumptions, and second we demonstrate detailed agreement with the recently proposed theory. These two goals are achieved by using data from both simulations and experiments. Finally, we show how measurements of the dipole fields pinpoint the theory parameters that determine the profile of the displacement field.

关键词： Elasticity Mechanical & acoustical properties Mechanical deformation Plasticity Amorphous materials

来源：评论

学校读者我要写书评

暂无评论

Noncooperative Model Predictive Game With Markov Jump Graph

引用

IEEE/CAA Journal of Automatica Sinica 2023年第4期10卷 931-944页

作者： Yang Xu Yuan Yuan Zhen Wang Xuelong Li School of Astronautics Northwestern Polytechnical UniversityXi’an 710072China School of Computer Science and Center for Optical Imagery Analysis and Learning Northwestern Polytechnical Universityand also with the School of CybersecurityNorthwestern Polytechnical UniversityXi’an 710072China School of Artificial Intelligence Optics and Electronics(iOPEN)Northwestern Polytechnical Universityand also with the Key Laboratory of Intelligent Interaction and ApplicationsNorthwestern Polytechnical UniversityMinistry of Industry and Information TechnologyXi’an 710072China

In this paper,the distributed stochastic model predictive control(MPC)is proposed for the noncooperative game problem of the discrete-time multi-player systems(MPSs)with the undirected Markov jump *** reflect the reality,the state and input constraints have been considered along with the external *** iterative algorithm is designed such that model predictive noncooperative game could converge to the socalledε-Nash equilibrium in a distributed *** conditions are established to guarantee the convergence of the proposed *** addition,a set of easy-to-check conditions are provided to ensure the mean-square uniform bounded stability of the underlying ***,a numerical example on a group of spacecrafts is studied to verify the effectiveness of the proposed method.

关键词： Markov jump graph model predictive control(MPC) multi-player systems(MPSs) noncooperative game ε-Nash equilibrium

来源：评论

学校读者我要写书评

暂无评论

Higher-order temporal interactions promote the cooperation in the multiplayer snowdrift game

引用

Science China(Information Sciences) 2023年第12期66卷 138-153页

作者： Yan XU Juan WANG Chengyi XIA Zhen WANG School of Computer Science and Engineering Tianjin University of Technology School of Electrical Engineering and Automation Tianjin University of Technology School of Artificial Intelligence Tiangong University School of Cyberspace and Center for OPTical IMagery Analysis and Learning(OPTIMAL) Northwestern Polytechnical University

To explore evolutionary dynamics of collective behaviors within the interconnected population,previous studies usually map non-pairwise interactions to higher-order static networks. However, from human communications to chemical reactions and biological systems, interactions often change over time, which cannot be simply described by higher-order static networks. In this study, we introduce time effects into higher-order networks and correspondingly investigate the evolutionary dynamics of multiplayer snowdrift games on higher-order temporal networks. Specifically, extensive simulations from four empirical datasets reveal that(1) the temporal effect of higher-order networks can facilitate the evolution of cooperation;(2) the higher-order topology can enhance the emergence of cooperation within a certain range of parameters;(3) the contribution of temporal burstiness and participants burstiness to cooperation is reversed. Furthermore,we theoretically demonstrate that the higher-order structure will suppress the propagation of defection in temporal networks. Our findings offer a new avenue for studying the evolution of altruistic behaviors in realistic complex networks.

关键词： collective behaviors non-pairwise interactions multiplayer snowdrift games temporal networks higher-order complex networks

来源：评论

学校读者我要写书评

暂无评论

Robust Semi-Supervised Deep Nonnegative Matrix Factorization with Constraint Propagation for Data Representation

引用

IEEE Transactions on Big Data 2025年

作者： Peng, Siyuan Yin, Jingxing Yang, Zhijing Nie, Feiping Chen, Badong Guangdong University of Technology School of Information Engineering 510006 China Northwestern Polytechnical University School of Computer Science Center for OPTical IMagery Analysis and Learning Xi'an710072 China Xi'an Jiaotong University Institute of Artificial Intelligence and Robotics Xi'an710049 China

Deep nonnegative matrix factorization (DNMF) technique has attached much great attention in recent year, since it can effectively discover the underlying hierarchical structure of complex data. However, most existing unsupervised and semi-supervised DNMF approaches not only suffer from the noisy data seriously, but also fail to enhance the decomposition quality of DNMF obviously by using the obtained supervisory information. To overcome these drawbacks, a robust semi-supervised DNMF method, called the correntropy based semi-supervised DNMF with constraint propagation (CSDCP), is proposed in this paper for learning a compact and meaningful data representation from the original data. Particularly, instead of adopting the traditional Frobenius norm, CSDCP employs the nonlinear and local similarity measure (e.g., correntropy) as the loss function in DNMF to enhance the robustness of DNMF for the noisy data. In addition, the hypergraph based constraint propagation (HCP) algorithm is adopted in CSDCP to exploit the limited supervisory information fully for capturing good data representation. Moreover, algorithm analysis of CSDCP is presented in this paper, including convergence analysis, robustness analysis supervised information analysis, and computational complexity. Extensive experimental results have illustrated that, in comparison to the most related DNMF approaches, CSDCP usually has better clustering results on six nonnegative datasets in clustering tasks. © 2015 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Triple discriminator generative adversarial network for zero-shot image classification

引用

Science China(Information Sciences) 2021年第2期64卷 5-18页

作者： Zhong JI Jiangtao YAN Qiang WANG Yanwei PANG Xuelong LI School of Electrical and Information Engineering Tianjin University Center for OPTical IMagery Analysis and Learning Northwestern Polytechnical University

One key challenge in zero-shot classification(ZSC) is the exploration of knowledge hidden in unseen classes. Generative methods such as generative adversarial networks(GANs) are typically employed to generate the visual information of unseen classes. However, the majority of these methods exploit global semantic features while neglecting the discriminative differences of local semantic features when synthesizing images, which may lead to sub-optimal results. In fact, local semantic information can provide more discriminative knowledge than global information can. To this end, this paper presents a new triple discriminator GAN for ZSC called TDGAN, which incorporates a text-reconstruction network into a dual discriminator GAN(D2GAN), allowing to realize cross-modal mapping from text descriptions to their visual representations. The text-reconstruction network focuses on key text descriptions for aligning semantic relationships to enable synthetic visual features to effectively represent images. Sharma-Mittal entropy is exploited in the loss function to make the distribution of synthetic classes be as close as possible to the distribution of real classes. The results of extensive experiments over the Caltech-UCSD Birds-2011 and North America Birds datasets demonstrate that the proposed TDGAN method consistently yields competitive performance compared to several state-of-the-art ZSC methods.

关键词： zero-shot classification generative adversarial nets text reconstruction Sharma-Mittal entropy

来源：评论

学校读者我要写书评

暂无评论

Anomalous elasticity and emergent dipole screening in three-dimensional amorphous solids

引用

Physical Review E 2023年第5期107卷 055005-055005页

作者： Harish Charan Michael Moshe Itamar Procaccia Department of Chemical Physics The Weizmann Institute of Science Rehovot 76100 Israel Racah Institute of Physics The Hebrew University of Jerusalem Jerusalem 9190 Israel Center for Optical Imagery Analysis and Learning Northwestern Polytechnical University Xi'an 710072 China

In recent work, we developed a screening theory for describing the effect of plastic events in amorphous solids on its emergent mechanics. The suggested theory uncovered an anomalous mechanical response of amorphous solids where plastic events collectively induce distributed dipoles that are analogous to dislocations in crystalline solids. The theory was tested against various models of amorphous solids in two dimensions, including frictional and frictionless granular media and numerical models of amorphous glass. Here we extend our theory to screening in three-dimensional amorphous solids and predict the existence of anomalous mechanics similar to the one observed in two-dimensional systems. We conclude by interpreting the mechanical response as the formation of nontopological distributed dipoles that have no analog in the crystalline defects literature. Having in mind that the onset of dipole screening is reminiscent of Kosterlitz-Thouless and hexatic transitions, the finding of dipole screening in three dimensions is surprising.

关键词： Elasticity Mechanical & acoustical properties Plasticity

来源：评论

学校读者我要写书评

暂无评论

Multitype View of Knowledge Contrastive learning for Recommendation

SSRN

引用

SSRN 2024年

作者： Zhang, Zhi-Hao Yang, Xiao-Jun Wu, Yang-Hui Wang, Jing Nie, Fei-Ping Guangdong University of Technology Guangzhou510006 China The School of Integrated Circuits Guangdong University of Technology Guangzhou510006 China The School of Computer Science The Center for Optical Imagery Analysis and Learning Northwestern Polytechnical University Xi’an710072 China

Graph Neural Networks (GNNs) are playing an increasingly vital role in the field of recommender systems. To improve knowledge perception within GNNs, contrastive learning has been applied and has proven to be highly effective. GNNs have the ability to aggregate diverse knowledge and integrate topologies, while contrastive learning seeks supervisory signals from the model data. The combination of GNNs and contrastive learning can improve recommendations. However, thoughtless or incomplete contrastive learning settings limit the effectiveness of GNNs-based recommender systems in learning knowledge from knowledge and interaction graphs. To better exploit the valuable information within knowledge graphs, we propose a novel multitype view of knowledge contrastive learning for recommendations (MVKC) *** MVKC model generates hierarchical views and augmented views in two modules, performing cross-hierarchicalview and cross-augmented-view contrastive learning and mining graph features in a self-supervised manner. The hierarchical views consist of global and local parts at multiple levels, while the augmented views are fused from the augmented knowledge graph and augmented interaction graph in our augmented processing. These features allow the MVKC model to alleviate the sparsity of user-item interaction graphs, suppress knowledge graph noise, and filter long-tail entities, which has been proven extremely important for a recommendation. The MVKC model also has strong anti-interference ability and robustness, which is crucial for a well-established model. Our experiments with three public datasets demonstrate that the MVKC model outperforms current state-of-the-art methods. © 2024, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Formation Control of Omnidirectional Mobile Robots Based on Bionic Coupling Mechanism

Formation Control of Omnidirectional Mobile Robots Based on ...

引用

2021 IEEE International Conference on Unmanned Systems, ICUS 2021

作者： Yang, Zhanwei Li, Shengjin Xu, Hao Yu, Dengxiu Wang, Zhen Philip Chen, C.L. Northwestern Polytechnical University School of Mechatronics Xian China Northwestern Polytechnical University Unmanned System Research Institute Xian China Northwestern Polytechnical University Center for Optical Imagery Analysis and Learning Xian China

ISBN: (纸本)9780738146577

In this paper, distributed formation control based on multiple omni-directional mobile robot systems(MOMRS) is studied. The model of omni-directional mobile robot exists the problems of nonlinear dynamics and input saturation, and there is a complicated coupling relationship in formation control. Firstly, according to the relationship between individuals in the natural community, a bionic coupling mechanism is proposed. Based on this, the formation control model of MOMRS with unknown nonlinear dynamics and input saturation constraints is established;Secondly, based on dynamic surface control and radial basis function neural network(RBF-NN), an adaptive formation controller is designed, which solves the problems of input saturation and nonlinear dynamics. Based on Lyapunov stability theorem, the stability of the controller is proved. Finally, six OMRs are used for simulation, and the simulation results prove the effectiveness of the designed controller. © 2021 IEEE.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

Creep failure of amorphous solids under tensile stress

引用

Physical Review E 2022年第3期106卷 034906-034906页

作者： Bhanu Prasad Bhowmik H. G. E. Hentschel Itamar Procaccia Department of Chemical Physics The Weizmann Institute of Science Rehovot 76100 Israel Department of Physics Emory University Atlanta Georgia 30322 USA Center for OPTical IMagery Analysis and Learning Northwestern Polytechnical University Xi'an 710072 China

Applying constant tensile stress to a piece of amorphous solid results in a slow extension, followed by an eventual rapid mechanical collapse. This “creep” process is of paramount engineering concern, and as such was the subject of study in a variety of materials, for more than a century. Predictive theories for τw, the expected time of collapse, are incomplete, mainly due to its dependence on a bewildering variety of parameters, including temperature, system size, tensile force, but also the detailed microscopic interactions between constituents. The complex dependence of the collapse time on all the parameters is discussed below, using simulations of strip of amorphous material. Different scenarios are observed for ductile and brittle materials, resulting in serious difficulties in creating an all-encompassing theory that could offer safety measures for given conditions. A central aim of this paper is to employ scaling concepts, to achieve data collapse for the probability distribution function (pdf) of lnτw. The scaling ideas result in a universal function which provides a prediction of the pdf of lnτw for out-of-sample systems, from measurements at other values of these parameters. The predictive power of the scaling theory is demonstrated for both ductile and brittle systems. Finally, we present a derivation of universal scaling function for brittle materials. The ductile case appears to be due to a plastic necking instability and is left for future research.

关键词： Creep Fracture Plasticity Rheology

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：