检索结果-内蒙古大学图书馆

2023 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2023

作者： Zhan, Lei Hu, Kai Zou, Longhao Yuan, Zhenhui Muntean, Gabriel-Miro Institue of Future Networks Southern University of Science and Technology Shenzhen China Peng Cheng Laboratory Department of Broadband Communication Shenzhen China Northumbria University Department of Computer and Information Sciences Newcastle United Kingdom Dublin City University School of Electronic Engineering Dublin Ireland

ISBN: (纸本)9798350321524

In the context of autonomous driving, moving objects such as vehicles and pedestrians are of critical importance as they primarily influence the maneuvering and braking of cars. Unfortunately, due to the limited detection range of sensors, some distant and blocked objects cannot be detected, leading to slow responses when some unexpected situations occur during driving. To address this problem, a real-time motion segmentation multi-task model (RMS), running on an individual vehicle, is introduced to provide motion segmentation of moving objects within its field of view. RMS consists of a shared encoder, a multi-modal fusion module, and a dual decoder. An enhanced High Definition (HD) map constructed with the proposed RMS in line with the recommendations of the 3rd Generation Partnership Project (3GPP) Vehicle-to-everything (V2X) communication standard is produced. Extensive experiments demonstrate how RMS outperforms existing state-of-the-art motion segmentation methods in terms of multiple metrics, including mean Intersection over Union (mIoU). Additionally, Internet of Vehicles (IoV) simulation experiments show how the time required to update the map is better than the times achieved when using other methods. © 2023 IEEE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Part-of-Speech Tagging Using Long Short Term Memory (LSTM): Amazigh Text Written in Tifinaghe Characters 6th

Part-of-Speech Tagging Using Long Short Term Memory (LSTM): ...

引用

6th International Conference on Business Intelligence, CBI 2021

作者： Maarouf, Otman El Ayachi, Rachid Laboratory TIAD Department of Computer Science Faculty of Science and Technology Sultan Moulay Slimane University Beni Mellal Morocco

ISBN: (纸本)9783030765071

Long short term memory (LSTM) networks have been gaining popularity in modeling sequential data such as phoneme recognition, speech translation, language modeling, speech synthesis, chatbot-like dialog systems, and others. This paper investigates the attention-based encoder-decoder LSTM networks in TIFINAGH part-of-speech (POS) tagging when it is compared to Conditional Random Fields (CRF) and Decision Tree. The attractiveness of LSTM networks is its strength in modeling long-distance dependencies. The experiment results show that Long short-term memory (LSTM) networks perform better than CRF and Decision Tree that have a near performance. © 2021, Springer Nature Switzerland AG.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

AI-based Gas Turbine Multi-Component Health Prognosis via Recurrent Expansion of Gas Path Parameters

AI-based Gas Turbine Multi-Component Health Prognosis via Re...

引用

2024 International Conference on Control, Automation and Diagnosis, ICCAD 2024

作者： Berghout, Tarek Amirat, Yassine Bounceur, Ahcene Benbouzid, Mohamed University of Batna 2 Laboratory of Automation and Manufacturing Engineering Batna Algeria Isen Yncréa Ouest L@bISEN Brest France King Fahd University of Petroleum & Minerals Department of Information and Computer Science Dhahran Saudi Arabia University of Brest Umr Cnrs 6027 Irdl Brest France

ISBN: (纸本)9798350361025

Investigating Gas Turbine (GT) degradation based on Gas Path Parameters (GPPs) is essential for its maintenance and operation. Monitoring GPPs is crucial for early detection of degradation, enabling timely maintenance interventions and preventing potential failures. Integrating physics-based thermodynamic models with representation learning significantly improves predictive maintenance studies and provides a solution to challenges posed by expensive and impractical accelerated aging experiments. However, this integration also presents challenges in managing both data complexity and data drift. In this context, this article aims to address these gaps by extending and improving previous research through (i) exclusive use of GPPs;(ii) the implementation of advanced data preprocessing techniques;and (iii) the use of innovative representation learning strategies. Specifically, it introduces ProgMachina, a tool for data quality analysis in prognosis studies that addresses issues related to data complexity in GPPs. Furthermore, to diversify the feature space and improve the adaptability of representation learning to degradation patterns, this paper proposes using Multiverse Recurrent Expansion with Multiple Repeats (MV-REMR) approach, which is based on a series of Recurrent Neural Networks (RNNs). For evaluation, this study incorporates cross-validation and multiple metrics while comparing against multiple RNNs and state-of-the-art works, demonstrating stable and promising performance, making it a suitable choice for GT prognosis tasks. © 2024 IEEE.

关键词： Gas turbines

来源：评论

学校读者我要写书评

暂无评论

Non-Line-of-Sight Multipath Classification Method for BDS Using Convolutional Sparse Autoencoder with LSTM

引用

Tsinghua science and Technology 2025年第1期30卷 68-86页

作者： Yahang Qin Zhenni Li Shengli Xie Bo Li Ming Liu Victor Kuzin School of Automation Guangdong University of Technologyand also with Guangdong-Hong Kong-Macao Joint Laboratory for Smart Discrete ManufacturingGuangzhou 510006China School of Automation Guangdong University of TechnologyGuangdong Key Laboratory of IoT Information TechnologyGuangzhou 510006China 111 Center for Intelligent Batch Manufacturing Based on IoT Technology Key Laboratory of Intelligent Information Processing and System Integration of IoTMinistry of EducationGuangzhou 510006China Department of Electronic and Computer Engineering Hong Kong University of Science and TechnologHong Kong 999077China Academician of Russian Engineering Academy Moscow Moscow 125009Russia

Multipath signal recognition is crucial to the ability to provide high-precision absolute-position services by the BeiDou Navigation Satellite System(BDS).However,most existing approaches to this issue involve supervised machine learning(ML)methods,and it is difficult to move to unsupervised multipath signal recognition because of the limitations in signal *** by an autoencoder with powerful unsupervised feature extraction,we propose a new deep learning(DL)model for BDS signal recognition that places a long short-term memory(LSTM)module in series with a convolutional sparse autoencoder to create a new autoencoder ***,we propose to capture the temporal correlations in long-duration BeiDou satellite time-series signals by using the LSTM module to mine the temporal change patterns in the time ***,we develop a convolutional sparse autoencoder method that learns a compressed representation of the input data,which then enables downscaled and unsupervised feature extraction from long-duration BeiDou satellite series ***,we add an l_(1/2) regularizer to the objective function of our DL model to remove redundant neurons from the neural network while ensuring recognition *** tested our proposed approach on a real urban canyon dataset,and the results demonstrated that our algorithm could achieve better classification performance than two ML-based methods(e.g.,11%better than a support vector machine)and two existing DL-based methods(e.g.,7.26%better than convolutional neural networks).

关键词： convolutional sparse autoencoder BeiDou Navigation Satellite System(BDS) long short-term memory(LSTM) multipath classification

来源：评论

学校读者我要写书评

暂无评论

ADEdgeDrop: Adversarial Edge Dropping for Robust Graph Neural Networks

arXiv

引用

arXiv 2024年

作者： Chen, Zhaoliang Wu, Zhihao Sadikaj, Ylli Plant, Claudia Dai, Hong-Ning Wang, Shiping Cheung, Yiu-Ming Guo, Wenzhong College of Computer and Data Science Fuzhou University Fuzhou350116 China Department of Computer Science Hong Kong Baptist University Hong Kong Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fuzhou350116 China Faculty of Computer Science and with the research network Data Science @ Uni Vienna University of Vienna Vienna1090 Austria

Although Graph Neural Networks (GNNs) have exhibited the powerful ability to gather graph-structured information from neighborhood nodes via various message-passing mechanisms, the performance of GNNs is limited by poor generalization and fragile robustness caused by noisy and redundant graph data. As a prominent solution, Graph Augmentation Learning (GAL) has recently received increasing attention in the literature. Among the existing GAL approaches, edge-dropping methods that randomly remove edges from a graph during training are effective techniques to improve the robustness of GNNs. However, randomly dropping edges often results in bypassing critical edges. Consequently, the effectiveness of message passing is weakened. In this paper, we propose a novel adversarial edgedropping method (ADEdgeDrop) that leverages an adversarial edge predictor guiding the removal of edges, which can be flexibly incorporated into diverse GNN backbones. Employing an adversarial training framework, the edge predictor utilizes the line graph transformed from the original graph to estimate the edges to be dropped, which improves the interpretability of the edge-dropping method. The proposed ADEdgeDrop is optimized alternately by stochastic gradient descent and projected gradient descent. Comprehensive experiments on eight graph benchmark datasets demonstrate that the proposed ADEdgeDrop outperforms state-of-the-art baselines across various GNN backbones, demonstrating improved generalization and robustness. Copyright © 2024, The Authors. All rights reserved.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Optical signatures of suppressed carrier localization in encapsulated WSe2monolayer

Optical signatures of suppressed carrier localization in enc...

引用

作者： Hossen, Raqibul Park, Sang-Hyuk Lee, Seong-Yeon Yee, Ki-Ju Yim, Sang-Youp Jho, Young-Dahl School of Electrical Engineering and Computer Science Gwangju Institute of Science and Technology Gwangju61005 Korea Republic of Gwangju Institute of Science and Technology Gwangju61005 Korea Republic of Clarendon Laboratory Department of Physics University of Oxford OxfordOX1 3PU United Kingdom Department of Physics Chungnam National University Daejeon34134 Korea Republic of

Low carrier mobility, closely associated with the formation of localized states, is the major bottleneck of utilizing the unique quantum transport properties in transition metal dichalcogenides (TMDCs). Here, we demonstrate an effective method to quantify the localization energy based on the temperature-dependent spectral variation of photoluminescence (PL) in pristine and hexagonal boron nitride (h-BN) encapsulated monolayer (ML) WSe2. Considering the protecting capability of h-BN against contamination and degradation, while not affecting the electronic structure as an insulating dielectric, the localization energy was comparatively extracted out of PL spectra in pristine and encapsulated ML WSe2. In pristine ML WSe2, two distinctive energy traces were resolved with an energy difference of about 17 meV, which was associated with the localized state revealed below 200 K. Clear evidence for the carrier localization was also evident in the integrated PL intensity trace with temperature as the trace from pristine ML clearly deviates from the dark-exciton-like behavior of ML WSe2, violating the spin selection rule of the lowest exciton state. In clear contrast, the temperature dependency of the h-BN encapsulated ML WSe2 in PL spectra matches well with the typical Varshni formula of free excitonic peaks and the integrated intensity trace of thermally populated spin subbands. Our study suggests that the h-BN encapsulation could suppress the carrier localization channels by avoiding surface oxidation due to air exposure and could provide insights into how one could preserve the excitonic features in TMDC materials and devices. © 2023 Author(s).

关键词： Boron nitride

来源：评论

学校读者我要写书评

暂无评论

Multilingual Lexical Simplification via Zero-Shot Paraphrasing

IEEE Transactions on Audio, Speech and Language Processing

引用

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 2200-2214页

作者： Kang Liu Jipeng Qiang Yun Li Yi Zhu Yunhao Yuan Kaixun Hua Xindong Wu Department of Computer Science Yangzhou University Yangzhou China Department of Industrial and Management Systems Engineering University of South Florida Tampa USA Key Laboratory of Knowledge Engineering with Big Data (the Ministry of Education of China) Hefei China

Lexical simplification (LS) method based on pretrained language models is a straightforward yet powerful approach for generating potential substitutes for a complex word through analysis of its contextual surroundings. Nonetheless, these methods necessitate distinct pretrained models tailored to diverse languages, often overlooking the imperative task of preserving a sentence’s meaning. In this paper, we propose a novel multilingual LS method via zero-shot paraphrasing (LSPG), as paraphrases provide diversity in word selection while preserving the sentence’s meaning. We regard paraphrasing as a zero-shot translation task within multilingual neural machine translation that supports hundreds of languages. Once the input sentence is channeled into the paraphrasing, we embark on the generation of the substitutes. This endeavor is underpinned by a pioneering decoding strategy that concentrates exclusively on the lexical modifications of the complex word. To utilize the strong capabilities of large language models (LLM), we further introduce a novel approach PromLS that incorporates the results of LSPG to generate heuristic-enhanced context, enabling the LLM to generate diverse candidate substitutions. Experimental results demonstrate that LSPG surpasses BERT-based methods and zero-shot GPT3-based methods significantly in English, Spanish, and Portuguese. We also demonstrate a substantial improvement achieved by PromLS compared to the previous state-of-the-art LLM approach. LS approaches usually assume that complex words and their replacements are individual terms, concentrating on word-for-word substitutions. To tackle the more challenging task of multi-word lexical simplification, including phrase-to-phrase replacements, we extend LSPG and PromLS into MultiLSPG and MultiPromLS. MultiLSPG identifies multi-word expressions matched with their corresponding word counts in specific positions, while MultiPromLS, akin to PromLS, utilizes these candidates to generate a heuristi

关键词： Multilingual Decoding Translation Rhetoric Large language models Training Neural machine translation Encoding Speech processing Probability distribution

来源：评论

学校读者我要写书评

暂无评论

An Improved Dissimilarity based Approach to Semantic Similarity Calculation 9

An Improved Dissimilarity based Approach to Semantic Similar...

引用

9th International Conference on Dependable Systems and Their Applications, DSA 2022

作者： Guan, Hui Ma, Tianyu Wang, Guangwei Shenyang University of Chemical Technology Department of Computer Science and Technology Liaoning ShenYang China Liaoning Key Laboratory of Intelligent Technology for Chemical Process Industry Liaoning Shenyang China

ISBN: (数字)9781665488778

ISBN: (纸本)9781665488778

With the trend that the computational process of semantic similarity more and more mimics the human thought process, it becomes very important to consider the difference between semantics. In this paper, we review commonly used semantic similarity models, discuss the limitations of our previous models, and improve the information content (IC) computation model based on wordnet ontology. In turn, different strategy is employed. The results show that the model proposed in this paper has high computational accuracy. © 2022 IEEE.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Multiple Redundant K-Means Clustered Scheme Based on Weighted Cluster Head Selection for Damaged S-UAV

Multiple Redundant K-Means Clustered Scheme Based on Weighte...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Grace Khayat Constandinos X Mavromoustakis Andreas Pitsillides Jordi Mongay Batalla Evangelos K. Markakis Department of Computer Science Mobile Systems Laboratory (MoSys Lab) University of Nicosia Nicosia Cyprus Department of Computer Science Cyprus and Department of Electrical & Electronic Engineering Science University of Cyprus University of Johannesburg (Visiting Professor) Warsaw University of Technology Warsaw Poland Department of Electrical and Computer Engineering Hellenic Mediteranean University Estavromenos Heraklion Greece

A swarm of unmanned aerial vehicles (S-UAVs) consists of UAVs flying together with the target of accomplishing a certain task in a faster and more reliable way as compared to a single UAV. In a crisis scenario, UAVs have been widely used in rescue missions. Clustering is one of the most reliable routing schemes for S-UAVs. The UAVs are grouped into clusters with a cluster-head (CH) and cluster-members (CM). The CH plays a major role in clustering schemes as it handles all inter-cluster communication. In a crisis case, any UAV is at risk of getting non-functional, thus resulting in a disconnected cluster. This paper proposes a new clustering scheme based on K-means and weighted formulas. The K-means protocol is applied to generate pilot phase clusters. Afterward, whenever the metrics of the networks are established, the weighted formula is applied for cluster formation and CH selection. The weighted formula is based on the performance index, the relative movement, and the remaining energy. To ensure end-to-end communication despite CH non-functionality, our proposed protocol selects a redundant CH for every CH. This protocol had been simulated using MATLAB. The results obtained and analyzed towards the end of this paper demonstrate that the proposed scheme is very promising.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Event-triggered leader-following formation control for multi-agent systems under communication faults: application to a fleet of unmanned aerial vehicles

引用

Journal of Systems Engineering and Electronics 2021年第5期32卷 1014-1022页

作者： VAZQUEZ TREJO Juan Antonio GUENARD Adrien ADAM-MEDINA Manuel PONSART Jean-Christophe CIARLETTA Laurent ROTONDO Damiano THEILLIOL Didier Research Center for Automatic Control of Nancy University of LorraineNancy F-54000France Electronic Engineering Department National Institute of Technology of MexicoCuernavaca Morelos 62490Mexico Lorraine Research Laboratory in Computer Science and Its Applications University of LorraineNancy F-54000France University of Stavanger Department of Electrical and Computer EngineeringStavanger 4021Norway

The main contribution of this paper is the design of an event-triggered formation control for leader-following consensus in second-order multi-agent systems(MASs)under communication *** the agents must follow the trajectories of a virtual leader despite communication faults considered as smooth time-varying delays dependent on the distance between the *** matrix inequalities(LMIs)-based conditions are obtained to synthesize a controller gain that guarantees stability of the synchronization *** on the closed-loop system,an event-triggered mechanism is designed to reduce the control law update and information exchange in order to reduce energy *** proposed approach is implemented in a real platform of a fleet of unmanned aerial vehicles(UAVs)under communication faults.A comparison between a state-of-the-art technique and the proposed technique has been provided,demonstrating the performance improvement brought by the proposed approach.

关键词： event-triggered leader-following consensus communication fault formation control unmanned aerial vehicle(UAV) experimental result

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：