检索结果-内蒙古大学图书馆

17th IEEE International conference on Mobility, Sensing and Networking (MSN)

作者： Reus-Muns, Guillem Salehi, Batool Roy, Debashri Jian, Tong Wang, Zifeng Dy, Jennifer Ioannidis, Stratis Chowdhury, Kaushik Northeastern Univ Dept Elect & Comp Engn Boston MA 02115 USA

ISBN: (纸本)9781665406680

Accurate beam alignment in the millimeter-wave (mmWave) band introduces considerable overheads involving brute-force exploration of multiple beam-pair combinations and beam retraining due to mobility. This cost becomes often intractable under high mobility scenarios, where fast beamforming algorithms that can quickly adapt the beam configurations are still under development for 5G and beyond. Besides, blockage prediction is a key capability in order to establish mmWave reliable links. In this paper, we propose a data fusion approach that takes inputs from visual edge devices and localization sensors to (i) reduce the beam selection overhead by narrowing down the search to a small set containing the best possible beam-pairs and (ii) detect blockage conditions between transmitters and receivers. We evaluate our approach through joint simulation of multi-modal data from vision and localization sensors and RF data. Additionally, we show how deep learning based fusion of images and Global Positioning System (GPS) data can play a key role in configuring vehicle-to-infrastructure (V2I) mmWave links. We show a 90% top-10 beam selection accuracy and a 92.86% blockage prediction accuracy. Furthermore, the proposed approach achieves a 99.7% reduction on the beam selection time while keeping a 94.86% of the maximum achievable throughput.

关键词： mmWave beam selection machine learning deep learning fusion multi-modal data 5G

来源：评论

学校读者我要写书评

暂无评论

Preliminary exploration of Geotechnical data Interpretation and GIS Integration: Toward AI Implementation

Preliminary Exploration of Geotechnical Data Interpretation ...

引用

Artificial Intelligence, Applications, Innovation and Ethics (AI2E), International conference for

作者： Danah S. Al-Enezi Kuwait Institute for Scientific Research Kuwait

ISBN: (数字)9798331513733

ISBN: (纸本)9798331513740

Kuwait's rapid urban expansion, driven by large-scale projects, such as AI-Mutlaa City and Silk City, highlights the need for comprehensive geotechnical data to support sustainable development. However, detailed studies on the subsurface geotechnical characteristics of the northern regions of Kuwait are lacking. This study focuses on establishing a preliminary geotechnical database, addressing challenges encountered during its development, and conducting a pilot analysis for AI-Mutlaa City. A questionnaire was distributed to geology, environmental, geotechnical, and engineering professionals to identify the challenges associated with creating the database. The survey revealed issues such as inconsistent data formats, limited access to borehole logs, and the hesitancy of private firms to share data due to security concerns. Despite these challenges, borehole data from ministries and private organizations were collected and organized using Microsoft Excel. These data formed the foundation of the geotechnical database, which was analyzed and visualized using Geographic Information System (GIS) tools. The pilot study in AI-Mutlaa City revealed predominantly dense soil layers, with significant variations in soil strength at a depth of 19.5 m, where standard penetration resistance ranged from 10 to 50 blows. These findings emphasize the variability of soil properties in the area and underscore the importance of site-specific investigations. The established geotechnical database provides valuable insights for construction planning, risk assessment, and geohazard management. Moving forward, the integration of artificial intelligence into the database is planned to enhance data processing, automate soil classification, and predict subsurface conditions in an effort to contribute to advanced geotechnical practices in Kuwait.

关键词： Surveys Technological innovation Accuracy databases Urban areas Soil properties visual databases Artificial intelligence Sustainable development Geographic information systems

来源：评论

学校读者我要写书评

暂无评论

Method for identifying geological inhomogeneities during surface seismic exploration based on finite difference numerical modeling 17

Method for identifying geological inhomogeneities during sur...

引用

17th conference and Exhibition Engineering and Mining Geophysics 2021

作者： Tsarev, R.I. Prigara, A.M. Voroshilov, V.A. JSC VNII Galurgii PSU

ISBN: (纸本)9789462823723

When performing surface seismic surveys in conditions of low contrast of acoustic properties of salt strata, visual interpretation does not work. The aim of the research is to establish relationships between the presence of geological heterogeneities and changes in the attributes of a seismic record. The basis for such studies is the need to create an objective and substantiated approach to the interpretation of seismic data in the study of a halogen formation. The work was carried out on the basis of seismic modeling, the initial data for which were the results of acoustic logging. The interpretation of the data was carried out by means of the attribute analysis of the synthetic sections obtained as a result of modeling. To optimize the process of working with seismic attributes, a special software AtAn has been developed, which allows performing correlation analysis using several calculation schemes. Based on the results of the attribute analysis of theoretical wave fields, obtained by us on the basis of solving the direct problem, a summary table was built, reflecting the effect of geological heterogeneities on the change in seismic attributes. © conference and Exhibition Eng. and Mining Geophysics 2021. All rights reserved.

关键词： Acoustic properties

来源：评论

学校读者我要写书评

暂无评论

Contrastive Initial State Buffer for Reinforcement Learning

Contrastive Initial State Buffer for Reinforcement Learning

引用

IEEE International conference on Robotics and Automation (ICRA)

作者： Nico Messikommer Yunlong Song Davide Scaramuzza Robotics and Perception Group Department of Informatics University of Zurich Switzerland Department of Neuroinformatics University of Zurich Switzerland ETH Zurich Switzerland

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

In Reinforcement Learning, the trade-off between exploration and exploitation poses a complex challenge for achieving efficient learning from limited samples. While recent works have been effective in leveraging past experiences for policy updates, they often overlook the potential of reusing past experiences for data collection. Independent of the underlying RL algorithm, we introduce the concept of a Contrastive Initial State Buffer, which strategically selects states from past experiences and uses them to initialize the agent in the environment in order to guide it toward more informative states. We validate our approach on two complex robotic tasks without relying on any prior information about the environment: (i) locomotion of a quadruped robot traversing challenging terrains and (ii) a quadcopter drone racing through a track. The experimental results show that our initial state buffer achieves higher task performance than the nominal baseline while also speeding up training convergence.

关键词： Training Legged locomotion Reinforcement learning Contrastive learning data collection Quadrupedal robots Task analysis

来源：评论

学校读者我要写书评

暂无评论

A Novel Clustering Algorithm with Dynamic Boundary Extraction Strategy Based on Local Gravitation 26th

A Novel Clustering Algorithm with Dynamic Boundary Extractio...

引用

26th Pacific-Asia conference on Knowledge Discovery and data Mining (PAKDD)

作者： Luo, Jiangmei Zhu, Qingsheng Li, Junnan Cheng, Dongdong Zhou, Mingqiang Chongqing Univ Coll Comp Sci Chongqing Peoples R China Yangtze Normal Univ Coll Big Data & Intelligent Engn Chongqing Peoples R China

ISBN: (纸本)9783031059360;9783031059353

Clustering has been widely used in visual analysis, pattern recognition, privacy protection and other fields. In recent years, numerous clustering methods have received increasing attention. However, discovering arbitrarily shaped clusters, determining the location and number of clustering cores and dealing with fuzzy boundaries is tough for most algorithms. We propose a novel clustering algorithm with dynamic boundary extraction strategy based on local gravitation (DBELG) which extracts boundary in a natural way, rather than mechanically defining a few core points. In order to identify fuzzy boundaries, a novel gravity model that makes use of three significant information about the data objects is proposed. The structure of the reserved core groups is clear and easy to cluster. On this basis, the core group clustering (CGC) is further proposed to cluster the core points. The experimental results show that DBELG achieves better performance than existing methods in handling datasets with fuzzy boundaries and complex structures.

关键词： Clustering Natural neighbor Boundary extraction Fuzzy boundaries Local gravitation

来源：评论

学校读者我要写书评

暂无评论

Disinformation analysis and Tracking Dashboard

Disinformation Analysis and Tracking Dashboard

引用

International conference on Military Communications and Information Systems (ICMCIS)

作者： Justina Mandravickaitė Milita Songailaitė Eglė Kankevičiūtė Anton Volčok Tomas Krilavičius Faculty of Informatics Vytautas Magnus University / CARD Lithuania Lingo data Lingo data / CARD Lithuania

The importance of disinformation identification and tracking in the information age is not questionable. War in Ukraine, information attacks in Baltic states, and a number of other disinformation attacks worldwide show that it is becoming a conventional tool in information space. In this paper, we present a disinformation analysis and tracking dashboard. It has 3 main components: Disinformation Analyzer, Match disinformation cases and Model Accuracy. The first component consists of a fined-tuned RoBERTa model for classifying textual input into fake and neutral. The second component contains semantic matching functionality for checking the semantic similarity of input text to the set of known disinformation cases. The third component includes the confusion matrix and model accuracy information. All 3 components have been combined into visual analytics platform for easy inspection and exploration of potential and known disinformation cases. The majority of the work was done during TIDE Hackathon 2023 as a part of the Disinformation Analyzer Challenge.

关键词：

来源：评论

学校读者我要写书评

暂无评论

visual Analytics for Profiling Land Use Changes

Visual Analytics for Profiling Land Use Changes

引用

Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI)

作者： Claudio Santos Maryam Hosseini João Rulff Fabio Miranda Luc Wilson Claudio Silva Nivan Ferreira Marcos Lage Universidade Federal Fluminense Massachusetts Institute of Technology New York University University of Illinois Chicago Universidade Federal de Pernambuco Kohn Pedersen Fox Associates PC

The growth of cities calls for regulations and zoning rules on how each piece of urban space will be used. Tracking land use can reveal a wealth of information about urban development. For that matter, cities have been releasing data sets describing the historical evolution of the shape and the attributes of land units. The complex nature of land-use data, however, makes the analysis of such data challenging and time-consuming. To address these challenges, we propose Urban Chronicles, a visual analytics system that enables interactive exploration of land-use changes. Using New York City's Primary Land Use Tax Lot Output (PLUTO), we show the system's capabilities to explore the data from several years at different scales. UrbanChron-iclessupports on-the-fly aggregation and filtering operations that leverage the hierarchical nature of the data set to index the shape and attributes of geographical regions that change over time. Finally, we demonstrate the system's utility through case studies that analyze the impact of Hurricane Sandy on land use attributes and the effects of rezoning plans in Brooklyn.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Interactive visual data exploration with subjective feedback: an information-theoretic approach

引用

data MINING AND KNOWLEDGE DISCOVERY 2020年第1期34卷 21-49页

作者： Puolamaki, Kai Oikarinen, Emilia Kang, Bo Lijffijt, Jefrey De Bie, Tijl Univ Helsinki Dept Comp Sci Helsinki Finland Univ Ghent Dept Elect & Informat Syst IDLab Ghent Belgium

visual exploration of high-dimensional real-valued datasets is a fundamental task in exploratory data analysis (EDA). Existing projection methods for data visualization use predefined criteria to choose the representation of data. There is a lack of methods that (i) use information on what the user has learned from the data and (ii) show patterns that she does not know yet. We construct a theoretical model where identified patterns can be input as knowledge to the system. The knowledge syntax here is intuitive, such as "this set of points forms a cluster", and requires no knowledge of maths. This background knowledge is used to find a maximum entropy distribution of the data, after which the user is provided with data projections for which the data and the maximum entropy distribution differ the most, hence showing the user aspects of data that are maximally informative given the background knowledge. We study the computational performance of our model and present use cases on synthetic and real data. We find that the model allows the user to learn information efficiently from various data sources and works sufficiently fast in practice. In addition, we provide an open source EDA demonstrator system implementing our model with tailored interactive visualizations. We conclude that the information theoretic approach to EDA where patterns observed by a user are formalized as constraints provides a principled, intuitive, and efficient basis for constructing an EDA system.

关键词： Exploratory data analysis Dimensionality reduction Information theory Subjective interestingness Maximum entropy distribution

来源：评论

学校读者我要写书评

暂无评论

An exploration of Machine Learning Methods for Gait analysis of Potential Guide Dogs 23

An Exploration of Machine Learning Methods for Gait Analysis...

引用

Proceedings of the Tenth International conference on Animal-Computer Interaction

作者： Yifan Wu Colt Nichols Marc Foster Devon Martin James Dieffenderfer Masataka Enomoto B. Duncan X. Lascelles Jane Russenberger Gerald Brenninkmeyer Alper Bozkurt David L. Roberts Dept of Electrical and Computer Engineering North Carolina State University USA College of Veterinary Medicine North Carolina State University USA International Working Dog Registry USA Guiding Eyes for the Blind USA Dept of Computer Science North Carolina State University USA

ISBN: (纸本)9798400716560

Gait analysis is important for guide dog organizations, as ideal guide dogs have a smooth and efficient gait, where they can also easily shift between and maintain various gaits. Gait quality and natural traveling speed are two of the multiple factors important in matching a guide dog to its visually impaired handler. Gait evaluation typically includes subjective visual observation of the dog or objective assessments obtained from special-designed equipment. Guide dog organizations need a method to easily collect and analyze objective data of gait information. In this work, we explored how various machine learning models could learn and analyze gait patterns from inertial measurements data that were collected during two different data collection experiments using a wearable sensor device. We also evaluated how well each machine learning model could generalize behavior patterns from various dogs under different environments. Additionally, we compared how sensor placement locations could affect gait prediction performance by attaching the sensor device to the dog’s neck and back area respectively. The tested machine learning models were able to classify different gaits in the range of 42% to 91% in terms of accuracy, and predict various gait parameters with an error rate ranging from 14% to 29% depending on the setup. Furthermore, we also observed that using behavior data collected from the neck region contains more movement information than the back area. By performing a cross-dataset generalization test on the machine learning models, we found that even with performance drop, the models were able to learn gait-specific behavior patterns that are generalizable for different dogs. Although the results were preliminary, the proposed gait analysis exploration still showed promising potential for studying behavior patterns of candidate guide dogs.

关键词： Canine Behavior analysis Canine Gait analysis Guide Dog IMU Machine Learning

来源：评论

学校读者我要写书评

暂无评论

A CNN-Based Framework for Underwater Image Noise Reduction and Visibility Enhancement in Marine Applications

A CNN-Based Framework for Underwater Image Noise Reduction a...

引用

International conference on Computer and Communications Technologies (ICCCT)

作者： Ashok P Ragul Sabarish J Vigneash R V Wasiq Rahman M A Antony Vijay J S. Kaliappan Dept of Computer Science and Engineering Sri Sairam Institute of Technology Chennai Dept of Computer Science and Engineering Karpagam College of Engineering Coimbatore Division of Research and Development Lovely Professional University Jalandhar

ISBN: (数字)9798331537579

ISBN: (纸本)9798331537586

Underwater imaging technology serves critical functions in diverse fields that range from marine biological research to underwater archaeological investigations to oceans exploration endeavors. The underwater imaging process typically produces highly degraded visual output due to multiple factors including light absorption and scattering while also affecting clarity because of particulate matter. We propose a Convolutional Neural Network (CNN)-based technique for underwater image noise reduction which works to enhance image visibility while improving perceptual data. Our system applies deep learning principles to master noise patterns while recovering image clarity. The proposed model trained using underwater images with different levels of noise underwent evaluation on the same dataset. Subjective visual assessment together with objective metrics demonstrated that the CNN delivered noise reduction performance at an accuracy range between 85-90% which defeated conventional noise filtering methods. The developed methodology exhibits substantial progress potential for seabed image quality enhancement tools which subsequently results in better underwater cognitive interpretation processes.

关键词： Image quality Deep learning visualization Recurrent neural networks Image color analysis Noise reduction Scattering Biology Convolutional neural networks Image enhancement

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：