检索结果-内蒙古大学图书馆

3rd International Workshop and Challenge on computer Vision in Endoscopy, EndoCV 2021

作者： Hong, Ayoung Lee, Giwan Lee, Hyunseok Seo, Jihyun Yeo, Doyeob Robotics Engineering Convergence Chonnam National University Gwangju Korea Republic of Department of AI Convergence Chonnam National University Gwangju Korea Republic of Daegu-Gyeongbuk Medical Innovation Foundation Daegu Korea Republic of Department of Computer Convergence Software Korea University Sejong Korea Republic of Electronics and Telecommunications Research Institute Daejeon Korea Republic of

Owing to the rapid development of deep learning technologies in recent years, autonomous diagnostic systems are widely used to detect abnormal lesions such as polyps in endoscopic images. However, the image characteristics, such as the contrast and illuminance, vary significantly depending on the center from which the data was acquired;this affects the generalization performance of the diagnostic method. In this paper, we propose an ensemble learning method based on k-fold cross-validation to improve the generalization performance of polyp detection and polyp segmentation in endoscopic images. Weighted box fusion methods were used to ensemble the bounding boxes obtained from each detection model trained for data from each center. The segmentation results of the data center-specific model were averaged to generate the final ensemble mask. We used a Mask R-CNN-based model for both the detection and segmentation tasks. The proposed method achieved a score of 0.7269 on the detection task and 0.7423 ± 0.2839 on the segmentation task in Round 1 of the EndoCV2021 challenge. © 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Probabilistic Surrogate networks for Simulators with Unbounded Randomness 38

Probabilistic Surrogate Networks for Simulators with Unbound...

引用

38th Conference on Uncertainty in Artificial Intelligence, Uai 2022

作者： Munk, Andreas Zwartsenberg, Berend Ścibior, Adam Baydin, Atılım Güneş Stewart, Andrew Fernlund, Goran Poursartip, Anoush Wood, Frank Department of Computer Science University of British Columbia Canada Inverted AI Ltd Canada Department of Engineering Science University of Oxford United Kingdom Convergent Manufacturing Technologies Inc Canada Composites Research Network University of British Columbia Canada Mila CIFAR AI Canada

ISBN: (纸本)9781713863298

We present a framework for automatically structuring and training fast, approximate, deep neural surrogates of stochastic simulators. Unlike traditional approaches to surrogate modeling, our surrogates retain the interpretable structure and control flow of the reference simulator. Our surrogates target stochastic simulators where the number of random variables itself can be stochastic and potentially unbounded. Our framework further enables an automatic replacement of the reference simulator with the surrogate when undertaking amortized inference. The fidelity and speed of our surrogates allow for both faster stochastic simulation and accurate and substantially faster posterior inference. Using an illustrative yet non-trivial example we show our surrogates' ability to accurately model a probabilistic program with an unbounded number of random variables. We then proceed with an example that shows our surrogates are able to accurately model a complex structure like an unbounded stack in a program synthesis example. We further demonstrate how our surrogate modeling technique makes amortized inference in complex black-box simulators an order of magnitude faster. Specifically, we do simulator-based materials quality testing, inferring safety-critical latent internal temperature profiles of composite materials undergoing curing. © 2022 Proceedings of the 38th Conference on Uncertainty in Artificial Intelligence, Uai 2022. All right reserved.

关键词： Random variables

来源：评论

学校读者我要写书评

暂无评论

Robust Positioning with LEO Satellites: Double-Difference Doppler Shift-Based Approach 100

Robust Positioning with LEO Satellites: Double-Difference Do...

引用

100th IEEE Vehicular Technology Conference, VTC 2024-Fall

作者： Hasan, Md. Ali Kabir, M. Humayun Islam, Md. Shafiqul Han, Sangmin Shin, Wonjae Ajou University Department of Artificial Intelligence Convergence Network Suwon Korea Republic of Islamic University Department of Electrical and Electronic Engineering Kushtia7003 Bangladesh Bangladesh University of Business and Technology Department of Computer Science and Engineering Dhaka Bangladesh Korea University School of Electrical Engineering Seoul Korea Republic of

ISBN: (纸本)9798331517786

Low Earth orbit (LEO) satellite-based positioning is emerging due to the available LEO constellation and high Doppler shift opportunity. However, positioning with LEO satellites utilizing signals of opportunity (SOPs)-based approach suffers from ephemeris error as the orbital element is estimated using two-line element (TLE) files that induce error over time. In the differential Doppler positioning method, an additional base receiver along with its known position with the user terminal (UT), is used to handle this erroneous measurement. The positioning accuracy of UT degrades significantly in the case of the long baseline (the distance between the base receiver and UT) scenario. Moreover, Doppler shift measurement errors also occur due to the clock synchronization issue between the base receiver and UT. Addressing these issues, we propose a robust positioning method, coined 3DPose, that can handle the clock synchronization issue between the base receiver and UT, as well as positioning degradation due to the long baseline. The proposed 3DPose method applies double-difference Doppler shift measurements to eliminate the clock synchronization issue, and a novel ephemeris error correction algorithm to improve UT positioning accuracy in case of the long baseline. To evaluate the performance of the proposed method, we perform a comparison with the existing differential Doppler positioning method. The comparison results show that the proposed 3DPose method outperforms the existing differential Doppler positioning in terms of positioning accuracy. © 2024 IEEE.

关键词： Orbits

来源：评论

学校读者我要写书评

暂无评论

RT-Swap: Addressing GPU Memory Bottlenecks for Real-Time Multi-DNN Inference 30

RT-Swap: Addressing GPU Memory Bottlenecks for Real-Time Mul...

引用

30th IEEE Real-Time and Embedded Technology and Applications Symposium, RTAS 2024

作者： Kang, Woosung Lee, Jinkyu Lee, Youngmoon Oh, Sangeun Lee, Kilho Chwa, Hoon Sung DGIST Dept. of Electrical Engineering and Computer Science Korea Republic of Dept. of Computer Science and Engineering Korea Republic of Hanyang University Dept. of Robotics Korea Republic of Ajou University Department of Software and Computer Engineering Korea Republic of School of AI Convergence Soongsil University Korea Republic of

ISBN: (纸本)9798350358414

The increasing complexity and memory demands of Deep Neural networks (DNNs) for real-Time systems pose new significant challenges, one of which is the GPU memory capacity bottleneck, where the limited physical memory inside GPUs impedes the deployment of sophisticated DNN models. This paper presents, to the best of our knowledge, the first study of addressing the GPU memory bottleneck issues, while simultaneously ensuring the timely inference of multiple DNN tasks. We propose RT-Swap, a real-Time memory management framework, that enables transparent and efficient swap scheduling of memory objects, employing the relatively larger CPU memory to extend the available GPU memory capacity, without compromising timing guarantees. We have implemented RT-Swap on top of representative machine-learning frameworks, demonstrating its effectiveness in making significantly more DNN task sets schedulable at least 72% over existing approaches even when the task sets demand up to 96.2% more memory than the GPU's physical capacity. © 2024 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Spatial throughput characterization in cognitive radio networks with threshold-based opportunistic spectrum access

Spatial throughput characterization in cognitive radio netwo...

引用

作者： Song, Xiaoshi Yin, Changchuan Liu, Danpu Zhang, Rui Beijing Key Laboratory of Network System Architecture and Convergence Beijing University of Posts and Telecommunications Beijing100876 China Department of Electrical and Computer Engineering National University of Singapore Singapore117576 Singapore

This paper studies the opportunistic spectrum access (OSA) of the secondary users in a large-scale overlay cognitive radio (CR) network. Two threshold-based OSA schemes, namely the primary receiver assisted (PRA) protocol and the primary transmitter assisted (PTA) protocol, are investigated. Under the PRA/PTA protocols, a secondary transmitter (ST) is allowed to access the spectrum only when the maximum signal power of the received beacons/pilots sent from the active primary receivers/transmitters (PRs/PTs) is lower than a certain threshold. To measure the resulting transmission opportunity for the secondary users by the proposed OSA protocols, the concept of spatial opportunity, which is defined as the probability that an arbitrary location in the primary network is detected as a spatial spectrum hole, is introduced and then evaluated by applying tools from stochastic geometry. Based on spatial opportunity, the coverage (non-outage transmission) performance in the overlay CR network is analyzed. With the obtained results of spatial opportunity and coverage probability, we finally characterize the spatial throughput, which is defined as the average spatial density of successful transmissions in the primary/secondary network, under the PRA and PTA protocols, respectively. © 1983-2012 IEEE.

关键词： Cognitive radio

来源：评论

学校读者我要写书评

暂无评论

Parallelized Jaccard-Based Learning Method and MapReduce Implementation for Mobile Devices Recognition from Massive network Data

引用

China Communications 2013年第7期10卷 71-84页

作者：刘军李银周 Felix Cuadrado Steve Uhlig 雷振明 Beijing Key Laboratory of Network System Architecture and Convergence Beijing University of Posts and Telecommunications Department of Electronic Engineering and Computer Science Queen MaryUniversity of London

The ability of accurate and scalable mobile device recognition is critically important for mobile network operators and ISPs to understand their customers' behaviours and enhance their user *** this paper,we propose a novel method for mobile device model recognition by using statistical information derived from large amounts of mobile network traffic ***,we create a Jaccardbased coefficient measure method to identify a proper keyword representing each mobile device model from massive unstructured textual HTTP access *** handle the large amount of traffic data generated from large mobile networks,this method is designed as a set of parallel algorithms,and is implemented through the MapReduce framework which is a distributed parallel programming model with proven low-cost and high-efficiency *** using real data sets show that our method can accurately recognise mobile client models while meeting the scalability and producer-independency requirements of large mobile network *** show that a 91.5% accuracy rate is achieved for recognising mobile client models from 2 billion records,which is dramatically higher than existing solutions.

关键词： mobile device recognition data mining Jaccard coefficient measurement distributed computing MapReduce

来源：评论

学校读者我要写书评

暂无评论

RSMA-Enabled Multigroup Multicast Rate-Matching for Multibeam Satellite Systems 24

RSMA-Enabled Multigroup Multicast Rate-Matching for Multibea...

引用

24th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2023

作者： Lee, Juhwan Seong, Jaehyup Lee, Jungwoo Le Ruyet, Didier Yin, Longfei Clerckx, Bruno Shin, Wonjae Seoul National University Department of Electrical and Computer Engineering Korea Republic of Ajou University Department of Artificial Intelligence Convergence Network Korea Republic of Conservatoire National des Arts et Métiers Cedric Laboratory France Imperial College Department of Electrical and Electronic Engineering London United Kingdom Austria Ajou University Department of Electrical and Computer Engineering Korea Republic of

ISBN: (纸本)9781665496261

Rate-splitting multiple access (RSMA) is a novel multiple access technique that breaks messages into common and private parts to mitigate interference in a flexible manner. Motivated by its robustness, RSMA is promising in multibeam satellite systems, where a satellite broadly serves terrestrial users through a multibeam multicast transmission with a full frequency reuse among all spot beams. In the multibeam satellite system, there are heterogeneous traffic demands among the spot beams. To meet the different traffic requirements for the spot beams, we present a novel RSMA-enabled rate-matching framework that matches the traffic demand for each spot beam through partially decoding inter-beam interference. We optimize the precoding and power allocation of the private/common messages to achieve the minimization of the difference between data rates and traffic demands among all spot beams. Numerical results show that the proposed framework significantly outperforms conventional transmission schemes in terms of meeting the traffic demands for the spot beams. © 2023 IEEE.

关键词： Multicasting

来源：评论

学校读者我要写书评

暂无评论

Risk Aware Trust of Information for Improving Safety in Vehicular networks

SSRN

引用

SSRN 2022年

作者： Choi, Jong-Young Ko, Young-Bae Department of AI Convergence Network Ajou University Suwon16499 Korea Republic of Department of Software and Computer Engineering Ajou University Suwon16499 Korea Republic of

In vehicular networks, safety applications based on basic safety messages (BSMs) can significantly improve road safety. BSMs broadcast by a vehicle contain real-time information on its kinematics, namely its position, speed, and heading. Outdated or incorrect BSMs can lead to collisions;therefore, BSM rate adaptation is necessary to communicate fresh and accurate information. Several recent studies have applied the age of information (AoI) concept to BSM rate adaptation. In this study, we investigate why "as is" application of AoI is not optimal and demonstrate the importance of selecting specific vehicles for AoI minimization. We also propose the novel trust of information (ToI) concept, which can be applied to quantify the reliability of BSMs. Through simulations, a ToI-based BSM rate-adaptation algorithm is compared with a recent BSM rate-adaptation scheme, demonstrating that the ToI algorithm outperforms the scheme in terms of safety assurance. © 2022, The Authors. All rights reserved.

关键词： Motor transportation

来源：评论

学校读者我要写书评

暂无评论

Towards Cache-Assisted Hierarchical Detection for Real-Time Health Data Monitoring in IoHT

引用

computers, Materials & Continua 2023年第11期77卷 2529-2544页

作者： Muhammad Tahir Mingchu Li Irfan Khan Salman AAl Qahtani Rubia Fatima Javed Ali Khan Muhammad Shahid Anwar School of Software Technology Dalian University of Technology(DUT)Dalian116621China School of Software Technology Key Laboratory for Ubiquitous Network and Service Software of Liaoning ProvinceDalian116650China Department of Computer Engineering King Saud UniversityRiyadh11543Saudi Arabia School of Software Tsinghua UniversityBeijing100084China Department of Software Engineering University of Science and TechnologyBanu28100Pakistan Department of AI and Software Gachon UniversitySeongnamsi13120Korea

Real-time health data monitoring is pivotal for bolstering road services’safety,intelligence,and efficiency within the Internet of Health Things(IoHT)***,delays in data retrieval can markedly hinder the efficacy of big data awareness detection *** advocate for a collaborative caching approach involving edge devices and cloud networks to combat *** strategy is devised to streamline the data retrieval path,subsequently diminishing network *** an adept cache processing scheme poses its own set of challenges,especially given the transient nature of monitoring data and the imperative for swift data transmission,intertwined with resource allocation *** paper unveils a novel mobile healthcare solution that harnesses the power of our collaborative caching approach,facilitating nuanced health monitoring via edge *** system capitalizes on cloud computing for intricate health data analytics,especially in pinpointing health *** the dynamic locational shifts and possible connection disruptions,we have architected a hierarchical detection system,particularly during *** system caches data efficiently and incorporates a detection utility to assess data freshness and potential lag in response ***,we introduce the Cache-Assisted Real-Time Detection(CARD)model,crafted to optimize *** the inherent complexity of the NP-hard CARD model,we have championed a greedy algorithm as a *** reveal that our collaborative caching technique markedly elevates the Cache Hit Ratio(CHR)and data freshness,outshining its contemporaneous benchmark *** empirical results underscore the strength and efficiency of our innovative IoHT-based health monitoring *** encapsulate,this paper tackles the nuances of real-time health data monitoring in the IoHT landscape,presenting a joint edge-cloud caching strategy paired with a hierarchical detection *** methodology yields enh

关键词： Real-time health data monitoring Cache-Assisted Real-Time Detection(CARD) edge-cloud collaborative caching scheme hierarchical detection Internet of Health Things(IoHT)

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning-Based Full-Duplex Communication UAV Base Station Trajectory Optimization in Disaster Environments

Deep Reinforcement Learning-Based Full-Duplex Communication ...

引用

2023 VTS Asia Pacific Wireless Communications Symposium, APWCS 2023

作者： Kim, Tae-Yoon Lee, Jaeyeol Kim, Jae-Hyun Ajou University Department of Artificial Intelligence Convergence Network Suwon16499 Korea Republic of Ajou University Department of Space Survey Information Technology Suwon16499 Korea Republic of Ajou University Department of Electrical and Computer Engineering Suwon16499 Korea Republic of

ISBN: (纸本)9798350316803

Unmanned aerial vehicles (UAVs) are widely used in various fields due to their fast mobility and flexible deployment capabilities. While previous research focused on using UAVs as relays, recent studies have explored the use of UAVs as base stations (UAV-BS) to overcome the limitations of ground-base stations. Particularly, in disaster environments where traditional ground-base stations are unavailable, the utilization of UAV-BS has gained significant attention for overcoming critical communication challenges. Moreover, research on full-duplex (FD) communication, which enhances frequency efficiency to serve a larger number of nodes in disaster environments, has been actively pursued. However, the availability of FD communication between node pairs can vary depending on the position of the UAV-BS when it is used. In this paper, we propose a deep reinforcement learning (DRL)-based approach to optimize the trajectory of UAV-BS for efficient FD communication in disaster environments. Our experimental results demonstrate that the proposed algorithm outperforms other existing algorithms, achieving more efficient FD communication within the given time constraints. © 2023 IEEE.

关键词： Base stations

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：