检索结果-内蒙古大学图书馆

Cross-domain object detection through coarse-to-fine feature adaptation

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zheng, Yangtao Huang, Di Liu, Songtao Wang, Yunhong Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University State Key Laboratory of Software Development Environment Beihang University School of Computer Science and Engineering Beihang University Beijing100191 China

Recent years have witnessed great progress in deep learning based object detection. However, due to the domain shift problem, applying off-the-shelf detectors to an unseen domain leads to significant performance drop. To address such an issue, this paper proposes a novel coarseto-fine feature adaptation approach to cross-domain object detection. At the coarse-grained stage, different from the rough image-level or instance-level feature alignment used in the literature, foreground regions are extracted by adopting the attention mechanism, and aligned according to their marginal distributions via multi-layer adversarial learning in the common feature space. At the fine-grained stage, we conduct conditional distribution alignment of foregrounds by minimizing the distance of global prototypes with the same category but from different domains. Thanks to this coarse-to-fine feature adaptation, domain knowledge in foreground regions can be effectively transferred. Extensive experiments are carried out in various cross-domain detection scenarios. The results are state-of-the-art, which demonstrate the broad applicability and effectiveness of the proposed approach. Copyright © 2020, The Authors. All rights reserved.

关键词： Object recognition

Can discovery of hidden charm strange pentaquark states help determine the spins of Pc(4440) and Pc(4457)?

学校读者我要写书评

暂无评论

Physical Review D 2021年第3期103卷 034003-034003页

作者： Ming-Zhu Liu Ya-Wen Pan Li-Sheng Geng School of Space and Environment Beihang University Beijing 100191 China School of Physics Beihang University Beijing 100191 China Beijing Key Laboratory of Advanced Nuclear Materials and Physics Beihang University Beijing 100191 China School of Physics and Microelectronics Zhengzhou University Zhengzhou Henan 450001 China Beijing Advanced Innovation Center for Big Data-Based Precision Medicine School of Medicine and Engineering Beihang University Beijing 100191 China

The pentaquark states Pc(4312), Pc(4440), and Pc(4457) could be nicely arranged into a multiplet of seven molecules of D¯(*)Σc(*) dictated by heavy quark spin symmetry. However, the spins of Pc(4440) and Pc(4457) are not yet fully determined. We employ the contact-range effective field theory to investigate the SU(3)-flavor counterparts of D¯(*)Σc(*) and study the possibility of whether their discovery could help determine the spins of Pc(4457) and Pc(4440). We find the existence of a complete hidden charm strange multiplet of D¯(*)Ξc(′)* molecules, irrespective of the spins of Pc(4440) and Pc(4457). On the other hand, we find that although molecules of D¯(*)Ξc are also likely, depending on the realization of the underlying dynamics, their discovery could be more useful for determining the spins of Pc(4440) and Pc(4457) and for telling us how heavy quark and light quark interactions depend on the spin of the light quark pair.

关键词： Particle interactions

Excited K meson, Kc(4180), with hidden charm as a DD¯K bound state

学校读者我要写书评

暂无评论

Physical Review D 2021年第3期103卷 L031501-L031501页

作者： Tian-Wei Wu Ming-Zhu Liu Li-Sheng Geng School of Physics Beihang University Beijing 102206 China School of Space and Environment Beihang University Beijing 102206 China Beijing Key Laboratory of Advanced Nuclear Materials and Physics Beihang University Beijing 100191 China Beijing Advanced Innovation Center for Big Data-Based Precision Medicine School of Medicine and Engineering Beihang University Beijing 100191 China School of Physics and Microelectronics Zhengzhou University Zhengzhou Henan 450001 China

Motivated by the recent discovery of two new states in the B+→D+D−K+ decay by the LHCb Collaboration, we study the DD¯K three-body system by solving the Schrödinger equation with the Gaussian Expansion Method. We show that the DD¯K system can bind with quantum numbers I(JP)=12(0−) and a binding energy of B3(DD¯K)=48.9−2.4+1.4 MeV. It can decay into J/ψK and DsD¯* via triangle diagrams, yielding a partial decay width of about 1 MeV. As a result, if discovered, it will serve as a highly nontrivial check on the nature of the many exotic hadrons discovered so far and on nonperturbative QCD as well. Assuming heavy quark spin symmetry, the same formalism is applied to study the DD¯*K system, which is shown to also bind with quantum numbers I(JP)=12(1−) and a binding energy of B3(DD¯*K)≃77.3−6.6+3.1 MeV, consistent with the results of previous works.

关键词： Chiral perturbation theory Hadron-hadron interactions Heavy quark effective theory Quark model

Multi-Channel Deep Networks on Sequence data for Multi-Action Recognition

学校读者我要写书评

暂无评论

Multi-Channel Deep Networks on Sequence Data for Multi-Actio...

International Conference on Intelligent Computation Technology and Automation, ICICTA

作者： Qianqian Qi Siwei Zhong Ye Wang Jun Long School of Computer Science and Engineering Central South University Changsha China Network Resource Management and Trust Evaluation Key Laboratory of Hunan Changsha China Big Data Institute Central South University Changsha China

Recently, Human Activity Recognition (HAR) using deep neural network has become a hot topic in human-computer interaction. Machine can effectively identify human naturalistic activities by learning from a large collection of sensor data. Distinguishing the same activity of different human is not only an interesting research problem, but also has real applications in private identity authentication. Based on the success of convolutional networks for sequence raw data and residual network to achieve the level of aesthetic representation of the automatic learning, we propose a novel model, named Deep Convolutional and Residual Networks (ConvResNets), for identifying authentication on wearable data, in which: (1)The raw input features can be replaced by high-level features representation; (2)ConvResNets does not require expert-knowledge to extract activity data features; (3)ConvResNets can easily capture features distribution of private activities. We evaluate our model on two public datasets. Our results show that our model is effective to recognize identity authentication via wearable datasets. We discuss the influence of networks parameters on performance to provide insights about its optimization.

关键词： Human computer interaction Deep learning Automation Computational modeling Authentication Feature extraction data models

Co-training Based on Semi-Supervised Ensemble Classification Approach for Multi-label data Stream

学校读者我要写书评

暂无评论

Co-training Based on Semi-Supervised Ensemble Classification...

IEEE International Conference on big knowledge (ICBK)

作者： Zhe Chu Peipei Li Xuegang Hu Key Laboratory of Knowledge Engineering with Big Data (Hefei University of Technology) Ministry of Education AnHui China School of Computer Science and Information Engineering Hefei University of Technology AnHui China

A large amount of data streams in the form of texts and images has been emerging in many real-world applications. These data streams often present the characteristics such as multi-labels, label missing and new class emerging, which makes the existing data stream classification algorithm face the challenges in precision space and time performance. This is because, on the one hand, it is known that data stream classification algorithms are mostly trained on all labeled single-class data, while there are a large amount of unlabeled data and few labeled data due to it is difficult to obtain labels in the real world. On the other hand, many of existing multi-label data stream classification algorithms mostly focused on the classification with all labeled data and without emerging new classes, and there are few semi-supervised methods. Therefore, this paper proposes a semi-supervised ensemble classification algorithm for multi-label data streams based on co-training. Firstly, the algorithm uses the sliding window mechanism to partition the data stream into data chunks. On the former w data chucks, the multi-label semi-supervised classification algorithm COINS based on co-training is used to training a base classifier on each chunk, and then an ensemble model with w COINS classifiers is generated ensemble model to adapt to the environment of data stream with a large number of unlabeled data. Meanwhile, a new class emerging detection mechanism is introduced, and the w+1 data chunk is predicted by the ensemble model to detect whether there is a new class emerging. When a new label is detected, the classifier is retrained on the current data chunk, and the ensemble model is updated. Finally, experimental results on five real data sets show that: as compared with the classical algorithms, the proposed approach can improve the classification accuracy of multi-label data streams with a large number of missing labels and new labels emerging.

关键词： data models Training Streaming media Adaptation models Partitioning algorithms Correlation knowledge engineering

Establishment of a novel engineering breeding technology for cytoplasmic male sterility

学校读者我要写书评

暂无评论

Crop Journal 2025年

作者： Xu, Na Yu, Dong Zhang, Longhui Wang, Tiankang Hu, Zhongxiao Lei, Quanqiang Qiu, Yingxin Yuan, Zheming Song, Shufeng Li, Li Hunan Engineering & Technology Research Centre for Agricultural Big Data Analysis & Decision-making College of Plant Protection Hunan Agricultural University Hunan Changsha 410128 China State Key Laboratory of Hybrid Rice Hunan Hybrid Rice Research Center Hunan Academy of Agricultural Sciences Hunan Changsha 410125 China Yuelushan Laboratory Hunan Changsha 410128 China Longping Branch College of Biology Hunan University Hunan Changsha 410125 China

The three-line hybrid rice system, which relies on cytoplasmic male sterility (CMS), faces challenges in breeding efficiency. To address these challenges, we developed CMS engineering breeding technology (CEBT). First, we constructed a complementation vector containing the fertility restoration module Double Rf4-Rf20, sorting marker gene DsRed2, and pollen lethal gene ZmAA1, then transformed this into a CMS line. We obtained reproductive lines without alterations to main agronomic traits, which can produce engineering sterile lines (ESL) and reproductive line seeds at a 1:1 ratio by self-crossing. CEBT converts the traditional three-line system into an efficient ‘new two-line’ hybrid framework: the reproductive line is improved while simultaneously improving the ESL, which significantly shortens the breeding process for three-line hybrid rice. © 2025 Crop Science Society of China and Institute of Crop Science, CAAS

关键词： CEBT CMS-WA Reproductive lines Rf gene Three-line hybrid rice

Optimization of user selection and bandwidth allocation for federated learning in VLC/RF systems

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Chuanhong Guo, Caili Yang, Yang Chen, Mingzhe Poor, H. Vincent Cui, Shuguang Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China Beijing Laboratory of Advanced Information Networks School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China Department of Electrical Engineering Princeton University PrincetonNJ08544 United States The Future Network of Intelligence Institute The Chinese University of Hong Kong Shenzhen518172 China Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen518172 China

Limited radio frequency (RF) resources restrict the number of users that can participate in federated learning (FL) thus affecting FL convergence speed and performance. In this paper, we first introduce visible light communication (VLC) as a supplement to RF in FL and build a hybrid VLC/RF communication system, in which each indoor user can use both VLC and RF to transmit its FL model parameters. Then, the problem of user selection and bandwidth allocation is studied for FL implemented over a hybrid VLC/RF system aiming to optimize the FL performance. The problem is first separated into two subproblems. The first subproblem is a user selection problem with a given bandwidth allocation, which is solved by a traversal algorithm. The second subproblem is a bandwidth allocation problem with a given user selection, which is solved by a numerical method. The final user selection and bandwidth allocation are obtained by iteratively solving these two subproblems. Simulation results show that the proposed FL algorithm that efficiently uses VLC and RF for FL model transmission can improve the prediction accuracy by up to 10% compared with a conventional FL system using only RF. Copyright © 2021, The Authors. All rights reserved.

关键词： Visible light communication

How Does knowledge Graph Embedding Extrapolate to Unseen data: A Semantic Evidence View

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Ren Cao, Yanan Zhu, Qiannan Bi, Guanqun Fang, Fang Liu, Yi Li, Qian Institute of Information Engineering Chinese Academy of Sciences China School of Cyber Security University of Chinese Academy of Sciences China Gaoling School of Artificial Intelligence Renmin University of China China Beijing Key Laboratory of Big Data Management and Analysis Methods China National Computer Network Emergency Response Technical Team Coordination Center of China China University of Technology Sydney Australia

knowledge Graph Embedding (KGE) aims to learn representations for entities and relations. Most KGE models have gained great success, especially on extrapolation scenarios. Specifically, given an unseen triple (h, r, t), a trained model can still correctly predict t from (h, r,?), or h from (?, r, t), such extrapolation ability is impressive. However, most existing KGE works focus on the design of delicate triple modeling function, which mainly tells us how to measure the plausibility of observed triples, but offers limited explanation of why the methods can extrapolate to unseen data, and what are the important factors to help KGE extrapolate. Therefore in this work, we attempt to study the KGE extrapolation of two problems: 1. How does KGE extrapolate to unseen data? 2. How to design the KGE model with better extrapolation ability? For the problem 1, we first discuss the impact factors for extrapolation and from relation, entity and triple level respectively, propose three Semantic Evidences (SEs), which can be observed from train set and provide important semantic information for extrapolation. Then we verify the effectiveness of SEs through extensive experiments on several typical KGE methods. For the problem 2, to make better use of the three levels of SE, we propose a novel GNN-based KGE model, called Semantic Evidence aware Graph Neural Network (SE-GNN). In SE-GNN, each level of SE is modeled explicitly by the corresponding neighbor pattern, and merged sufficiently by the multi-layer aggregation, which contributes to obtaining more extrapolative knowledge representation. Finally, through extensive experiments on FB15k-237 and WN18RR datasets, we show that SE-GNN achieves state-of-the-art performance on knowledge Graph Completion task and performs a better extrapolation ability. Our code is available at https://***/renli1024/SE-GNN. Copyright © 2021, The Authors. All rights reserved.

关键词： knowledge graph

Unsupervised Domain Adaptive Person Re-Identification via Human Learning Imitation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Peng, Yang Liu, Ping Luo, Yawei Zhou, Pan Xu, Zichuan Liu, Jingen Center for Frontier AI Research Agency for Science Technology and Research Singapore Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Huazhong University of Science and Technology Hubei Wuhan China College of Computer Science and Technology Zhejiang University Zhejiang Hangzhou China Key Laboratory for Ubiquitous Network and Service Software of Liaoning Province School of Software Dalian University of Technology Liaoning Dalian China

Unsupervised domain adaptive person re-identification has received significant attention due to its high practical value. In past years, by following the clustering and finetuning paradigm, researchers propose to utilize the teacher-student framework in their methods to decrease the domain gap between different person re-identification datasets. Inspired by recent teacher-student framework based methods, which try to mimic the human learning process either by making the student directly copy behavior from the teacher or selecting reliable learning materials, we propose to conduct further exploration to imitate the human learning process from different aspects, i.e., adaptively updating learning materials, selectively imitating teacher behaviors, and analyzing learning materials structures. The explored three components, collaborate together to constitute a new method for unsupervised domain adaptive person re-identification, which is called Human Learning Imitation framework. The experimental results on three benchmark datasets demonstrate the efficacy of our proposed method. Copyright © 2021, The Authors. All rights reserved.

关键词： Students