检索结果-内蒙古大学图书馆

TechRxiv 2023年

作者： Hao, Lifei Huang, Baoqi Jia, Bing Mao, Guoqiang The Engineering Research Center of Ecological Big Data Ministry of Education The Inner Mongolia A.R. Key Laboratory of Wireless Networking and Mobile Computing The College of Computer Science Inner Mongolia University Hohhot010021 China The Research Institute of Smart Transportation Xidian University Xi’an710071 China

Regarding the passive WiFi sensing based crowd analysis, this paper first theoretically investigates its limitations, and then proposes a deep learning based scheme targeted for returning fine-grained crowd states in large surveillance areas. To this end, three key challenges are coped with: to relieve the influences of the randomness and sparsity induced by passive WiFi sensing, an attention-based deep convolutional autoencoder model is designed to recover accurate crowd density maps in a way similar to image reconstruction;to combat the anonymity caused by MAC randomization, following the identification of local high-density crowds (LHDCs) with the density clustering algorithm, i.e. DM-DBSCAN, a bidirectional convolutional LSTM based model is employed to infer LHDC speeds;to overcome the absence of passive WiFi sensing datasets for model training, three semi-synthetic datasets are produced by emulating passive WiFi sensing with practical pedestrian tracking datasets. Extensive experiments confirm that, the proposed scheme significantly outperforms existing WiFi-based methods in terms of crowd density estimation and provides superior crowd speed estimation. More importantly, the scheme can also produce consistent crowd states on a real-world dataset, revealing that it has the ability to support accurate, visualized and real-time crowd monitoring in large surveillance areas. © 2023, CC BY-NC-SA.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Classification-based self-learning for weakly supervised bilingual lexicon induction 58

Classification-based self-learning for weakly supervised bil...

引用

58th Annual Meeting of the Association for Computational Linguistics, ACL 2020

作者： Karan, Mladen Vulić, Ivan Korhonen, Anna Glavaš, Goran TakeLab Faculty of Electrical Engineering and Computing University of Zagreb Language Technology Lab TAL University of Cambridge Data and Web Science Group University of Mannheim

ISBN: (纸本)9781952148255

Effective projection-based cross-lingual word embedding (CLWE) induction critically relies on the iterative self-learning procedure. It gradually expands the initial small seed dictionary to learn improved cross-lingual mappings. In this work, we present CLASSYMAP, a classification-based approach to self-learning, yielding a more robust and a more effective induction of projection-based CLWEs. Unlike prior self-learning methods, our approach allows for integration of diverse features into the iterative process. We show the benefits of CLASSYMAP for bilingual lexicon induction: we report consistent improvements in a weakly supervised setup (500 seed translation pairs) on a benchmark with 28 language pairs. © 2020 Association for Computational Linguistics

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Transformer-Based Hierarchical Dynamic Decoders for Salient Object Detection

SSRN

引用

SSRN 2023年

作者： Zheng, Qingping Zheng, Ling Deng, Jiankang Li, Ying Shang, Changjing Shen, Qiang School of Computer Science National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Shaanxi Provincial Key Laboratory of Speech & Image Information Processing Northewestern Polytechnical University Shaanxi Xi’an710129 China Fuzhou Institute of Data Technology Fujian350200 China Department of Computing Imperial College London LondonSW7 2AZ United Kingdom

HighlightsTransformer-Based Hierarchical Dynamic Decoders for Salient Object DetectionQingping Zheng, Ling Zheng, Jiankang Deng, Ying Li, Changjing Shang, Qiang Shen• T-HDDNet employs dynamic upsampling and fusion for data-driven salient object detection.• T-HDDNet uses a spatially sensitive loss to learn salient object structures, yielding a high-resolution map.• Our method exceeds benchmarks at 22.19G MACs and 27M Params, reducing costs compared to the top model. © 2023, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

E-Pdo: An Enhanced Prairie Dog Optimization Algorithm with ĹEvy Distribution and Dynamic Opposition-Based Learning for Global Optimization and Engineering Problems

SSRN

引用

SSRN 2023年

作者： Biswas, Saptadeep Shaikh, Azharuddin Ezugwu, Absalom El-Shamir Mirjalili, Seyedali Bera, Uttam Kumar Department of Mathematics National Institute of Technology Tripura Agartala India Unit for Data Science and Computing North-West University 11 Hoffman Street Potchefstroom2520 South Africa Centre for Artificial Intelligence Research and Optimization Torrens University Australia Fortitude Valley BrisbaneQLD4006 Australia

As a recent meta-heuristic algorithm, the Prairie Dog Optimization (PDO) algorithm mimics the foraging and borrowingbehaviour of prairie dogs. PDO is very dynamic, but premature convergence occurs. Convergence is slow because of thecumulative sum of the random walk, which has a long running time. This paper attempts to change the implementationstructure to improve performance. We use the Ĺevy dynamic random walk technique and a dynamic opposite learningstrategy (DOL) with a modified population initialization and an improved generation-jumping strategy to develop a newversion of the PDO (e-PDO). The DOL scheme enhances the process of reducing the chances of premature *** increases population diversity by avoiding a local optimum. It also covers global optimization and operationsfunctions. Modified Ĺevy flight with random walk helps reduce the running time when the e-PDO finds the optimalvalue of the test function. We analyse 33 typical benchmark issues from CEC 2017 and evaluate the performance of theproposed approach against seven comparative techniques (GWO, MFO, ALO, WOA, DA, SCA, and RSA). The proposede-PDO algorithm, according to numerical results, produces good outcomes. In addition, to validate the e-PDO resultsand reliability, we use average rank tests, measurement of alternatives and ranking according to the compromise solution(MARCOS) method and convergence report of the e-PDO and other algorithms. Moreover, two design problems aresolved by the proposed e-PDO algorithm. According to the results, we achieved very impressive results, indicating thatthe algorithm is well suited to solving practical problems. © 2023, The Authors. All rights reserved.

关键词： Swarm intelligence

来源：评论

学校读者我要写书评

暂无评论

Emergent competition shapes the ecological properties of multi-trophic ecosystems

arXiv

引用

arXiv 2023年

作者： Feng, Zhijie Marsland, Robert Rocks, Jason W. Mehta, Pankaj Department of Physics Boston University BostonMA02215 United States Pontifical University of the Holy Cross Rome Italy Biological Design Center Boston University BostonMA02215 United States Faculty of Computing and Data Science Boston University BostonMA02215 United States

Ecosystems are commonly organized into trophic levels – organisms that occupy the same level in a food chain (e.g., plants, herbivores, carnivores). A fundamental question in theoretical ecology is how the interplay between trophic structure, diversity, and competition shapes the properties of ecosystems. To address this problem, we analyze a generalized Consumer Resource Model with three trophic levels using the zero-temperature cavity method and numerical simulations. We find that intra-trophic diversity gives rise to "emergent competition" between species within a trophic level due to feedbacks mediated by other trophic levels. This emergent competition gives rise to a crossover from a regime of top-down control (populations are limited by predators) to a regime of bottom-up control (populations are limited by primary producers) and is captured by a simple order parameter related to the ratio of surviving species in different trophic levels. We show that our theoretical results agree with empirical observations, suggesting that the theoretical approach outlined here can be used to understand complex ecosystems with multiple trophic levels. © 2023, CC BY.

关键词： Ecosystems

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning Based Control Domain Division in LEO Satellite Networks

Reinforcement Learning Based Control Domain Division in LEO ...

引用

IEEE International Conference on High Performance computing and Communications (HPCC)

作者： Feilong Tang Xue Li Long Chen Jiacheng Liu Ming Gao Yanqin Yang Wenchao Xu Heteng Zhang School of Data Science and Engineering East China Normal University Shanghai China School of Computing Science Simon Fraser University Vancouver Canada Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong China School of Communication and Electric Engineering East China Normal University Shanghai China Suzhou All-Space-Time Information Technology Co. Ltd. Suzhou China

The performance of SDN-based LEO satellite networks significantly depends on the control domain division approaches. The dynamical topology in LEO networks results in the time-varying delay for network management. Therefore, the control domain division is very challenging and should be dynamical. In this paper, we propose a control domain division approach based on reinforcement learning (RL), which aims at reducing the average management delay of controllers during the control period, and as well as minimizing the number of control domain handovers. Firstly, we analyze the factors that need to be paid attention to in controller deployment, such as delay, load and controller switching cost, and formulate the controller deployment. By relaxing the constraints, we prove that the problem is NP-hard. Secondly, we propose an approach that considers future motion trajectory based on reinforcement learning (FMT _ RL) to deploy controllers. Since the dimension of the neural network in reinforcement learning is fixed, we use the improved K-Means algorithm to select the set of mechanisms to determine the dimension of the neural network. Particularly, in order to obtain the node composition control domain, the observation including both future motion trajectory and history controller selection information are input into the neural network to calculate the control domain division results. The experimental results demonstrate that our approach significantly outperforms related approaches, with better stability and average control cost.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Human-Computer Interface Challenges in Mobile data Mining

Human-Computer Interface Challenges in Mobile Data Mining

引用

Sustainable computing and Smart Systems (ICSCSS), International Conference on

作者： K. Padmanaban A. Sathiya K. Jeevitha R. Anto Arockia Rosaline Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vijayawada Andhra Pradesh India Department of Artificial Intelligence and Data Science Sri Sairam Institute of Technology Chennai Tamil Nadu India Department of Electronics and Communication Engineering R.M.K. Engineering College Chennai Tamil Nadu India Department of Computing Technologies SRM Institute of Science and Technology Chennai Tamil Nadu India

The continuous advancement of computer software, hardware, and artificial intelligence, has allowed for tremendous progress in the development of human-machine communication systems in recent years. The development of such methods provides more accurate results than humans. The core of the human-computer interface is the Natural Language Understanding (NLU) component. The success of Human-Computer Interaction (HCI) and the time it takes to accomplish tasks are both directly affected by the results of this module’s semantic comprehension. In this research study, a neural network and HCI-based engine is developed for understanding and interacting with natural languages. data from the appropriate literature is used to comprehend the necessary understanding of HCI. Next, the interactive engine is used to develop a model of HCI based on natural language comprehension using a neural network, and also the model’s underlying neural network reasoning structure is assessed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A data-driven analysis on the mediation effect of compartment models between control measures and COVID-19 epidemics

arXiv

引用

arXiv 2023年

作者： Zhang, Dongyan Yang, Wuyue Wen, Wanqi Peng, Liangrong Zhuge, Changjing Hong, Liu School of Mathematics Sun Yat-Sen University Guangdong Guangzhou510275 China Beijing Institute for Scientific and Engineering Computing Department of Mathematics Faculty of Science Beijing University of Technology Beijing100124 China Yanqi Lake Beijing Institute of Mathematical Sciences and Applications Beijing101408 China College of Mathematics and Data Science Minjiang University Fujian Fuzhou350108 China

Background On May 5th, 2023, WHO declared an end to the global COVID-19 public health emergency, which means a significant transition from global critical emergency response activities to long-term sustained COVID-19 prevention and control. At this very moment, we make a retrospective review on various control measures taken by 127 countries/territories during the first wave of COVID-19 pandemic until July 7, 2020, and evaluate their impacts on the epidemic dynamics quantitatively. Methods We collect and evaluate the control measures implemented by each country. The SEIR-QD model, as a representative for general compartment models, is used to fit the epidemic data, enabling the extraction of crucial model parameters and dynamical features. The mediation effect of the SEIR-QD model is revealed by using the mediation analysis with structure equation modeling for multiple mediators operating in parallel. The inherent impacts of these control policies on the transmission dynamics of COVID-19 epidemics are clarified and compared with results derived from both multiple linear regression and neural-network-based nonlinear regression. Results A dramatic distinction in the control measures is observed among major countries/territories around the world, which largely affects the spreading rate and infected population size in each country. Several key dynamical features, like the normalized cumulative numbers of confirmed/cured/death cases on the 100th day and the half time, show statistically significant linear correlations with the control measures, thereby confirming the dramatic impacts of control measures and their respective implementation intensity on the epidemic. Most importantly, the SEIR-QD model, especially its infection rate and protection rate, has been confirmed to exhibit a statistical significant mediation effect between the control measures and dynamical features of epidemics. The mediation effect along the pathway from control measures in Category 2 to four

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Generation of Dynamic Authorization Using Cryptography For Private-Keyless data Sharing

SSRN

引用

SSRN 2023年

作者： Lu, Hai Zhu, Yan Yau, Stephen S. Zhu, Guizhen Lv, Kewei School of Computer and Communication Engineering University of Science and Technology Beijing Beijing100083 China School of Computing Informatics and Decision Systems Engineering Arizona State University TempeAZ85281 United States Data and Communication Science Technology Research Institute Beijing100191 China School of Cyber Security University of Chinese Academy of Sciences Beijing100093 China

Private-keyless access is emerging as an inevitable trend in the evolution of future information systems. It helps simplify key management, enhance security, and increase flexibility of access authorization. Hereby, we introduce a novel security model, Cryptographic Resource Enabled Infrastructure (CREI), to support private-keyless access and policy-based data custody. Firstly, in the data custody phase, data resource and its session key are uploaded in ciphertext form to CREI, and the access authorization complies with multiple policies designated by the provider. Secondly, in the dynamic authorization phase, traditional keys or keycards are replaced with real-time attribute credentials issued by multiple authorities. CREI makes dynamically cryptographic policy decision according to the credentials and the cryptographic policy (cryptopolicy). Technically, we design the Credentials-Driven Cryptosystem (CDC) over ideal lattice to build anti-quantum confidence for CREI. In CDC, preimage Gaussian sampler is adopted to sample short vector as real-time attribute credential instead of user's private key. Small Policy Matrix (SPM) is implemented to convert monotone policies into optimized cryptopolicies in order to reduce accumulated error. The security analysis confirms that the credentials, cryptopolicy and its decision result are existentially unforgeable, ensuring the semantic security of the whole CREI. Experimental results indicate CREI has lower storage and computational costs than existing schemes. Meanwhile, despite continuous improvements to the access policies as system evolves, CREI can simply select an appropriate policy among multiple candidates to make authorization decisions without having to re-encrypt data. © 2023, The Authors. All rights reserved.

关键词： Authorization

来源：评论

学校读者我要写书评

暂无评论

Self-supervised Image Clustering from Multiple Incomplete Views via Constrastive Complementary Generation

arXiv

引用

arXiv 2022年

作者： Wang, Jiatai Xu, Zhiwei Yang, Xuewen Guo, Dongjin Liu, Limin College of Data Science and Application Inner Mongolia University of Technology Huhhot010080 China Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China InnoPeak Technology Inc CA94303 United States

Incomplete Multi-View Clustering aims to enhance clustering performance by using data from multiple modalities. Despite the fact that several approaches for studying this issue have been proposed, the following drawbacks still persist: 1) It’s difficult to learn latent representations that account for complementarity yet consistency without using label information;2) and thus fails to take full advantage of the hidden information in incomplete data results in suboptimal clustering performance when complete data is scarce. In this paper, we propose Contrastive Incomplete Multi-View Image Clustering with Generative Adversarial Networks (CIMIC-GAN), which uses GAN to fill in incomplete data and uses double contrastive learning to learn consistency on complete and incomplete data. More specifically, considering diversity and complementary information among multiple modalities, we incorporate autoencoding representation of complete and incomplete data into double contrastive learning to achieve learning consistency. Integrating GANs into the autoencoding process can not only take full advantage of new features of incomplete data, but also better generalize the model in the presence of high data missing rates. Experiments conducted on four extensively-used datasets show that CIMIC-GAN outperforms state-of-the-art incomplete multi-View clustering methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Generative adversarial networks

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：