检索结果-内蒙古大学图书馆

34th International Florida Artificial Intelligence Research Society Conference, FLAIRS-34 2021

作者： Badr, Haidi Wanas, Nayer Fayek, Magda Electronics Research Institute Informatics Dapartment Egypt Cairo university Faculty of Engineering Computer Department Egypt

Since labeled data availability differs greatly across domains, Domain Adaptation focuses on learning in new and unfamiliar domains by reducing distribution divergence. Recent research suggests that the adversarial learning approach could be a promising way to achieve the domain adaptation objective. Adversarial learning is a strategy for learning domain-transferable features in robust deep networks. This paper introduces the TSAL paradigm, a two-step adversarial learning framework. It addresses the real-world problem of text classification, where source domain(s) has labeled data but target domain (s) has only unlabeled data. TSAL utilizes joint adversarial learning with class information and domain alignment deep network architecture to learn both domain-invariant and domain-specific features extractors. It consists of two training steps that are similar to the paradigm, in which pre-trained model weights are used as initialization for training with new data. TSAL’s two training phases, however, are based on the same data, not different data, as is the case with fine-tuning. Furthermore, TSAL only uses the learned domain-invariant feature extractor from the first training as an initialization for its peer in subsequent training. By doubling the training, TSAL can emphasis the leverage of the small unlabeled target domain and learn effectively what to share between various domains. A detailed analysis of many benchmark datasets reveals that our model consistently outperforms the prior art across a wide range of dataset distributions. © 2021by the authors. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Predictive Design for Quality Assessment Employing Cloud Computing And Machine Learning 3

Predictive Design for Quality Assessment Employing Cloud Com...

引用

3rd International Conference on Advance Computing and Innovative Technologies in engineering, ICACITE 2023

作者： Jindal, Gaurav Tiwari, Vidhika Mahomad, Riyaz Gehlot, Anita Jindal, Mimoh Bordoloi, Dibyhash G L Bajaj College of Technology and Management Department of Master of Computer Applications Gr. Noida India DPGITM Mechanical Engineering Department Gurugram India SR University School of Computer Science and Artificial Intelligence Telangana Warangal India Uttaranchal University Uttaranchal Institute of Technology India National Informatics Centre Delhi India Graphic Era Deemed to be University Department of Computer Science & Engineering Uttarakhand Dehradun248002 India

ISBN: (纸本)9798350399264

In this work, we investigate a unique effective framework for projected design inspections in industrial manufacturing combining machine learning methodologies and edges cloud computing technologies. We recommend a thorough approach that involves targeted data gathering and broadcast, forecasting and supplying appropriate, as well as innovation with the current IT plant infrastructure, in contrary to state-of-the-art contributions. To highlight the steps and advantages of the suggested solution, a genuine business use case in the manufacture of SMT is described. The outcomes demonstrate that the proposed strategy can dramatically reduce inspection volumes, leading to economic gains. A key achievement basis for the drawn-out presentation of assembling ventures is the creation of imperfection-free, top-notch items. Indeed, even with the intricacy and assortment of items and the requirement for savvy production, an exhaustive and dependable quality investigation is habitually required. Along these lines, high examination volumes cause producing bottlenecks in the review processes. Thus, the purpose of this paper is to develop a predictive design for quality assessment regarding employing cloud computing and machine learning. Concerning this particular research, a descriptive research design has been used by depending on secondary data due to its reliability and validity. Moreover, it has been found that both cloud computing and machine learning contribute effectively for employing quality assessment because of their innovative features and infrastructure. © 2023 IEEE.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Enhancing EEG Signal-Based Emotion Recognition with Synthetic Data: Diffusion Model Approach

arXiv

引用

arXiv 2024年

作者： Siddhad, Gourav Iwamura, Masakazu Roy, Partha Pratim Department of Computer Science and Engineering Indian Institute of Technology Uttarakhand Roorkee247667 India Department of Core Informatics Graduate School of Informatics Osaka Metropolitan University Sakai Osaka599-8531 Japan

Emotions are crucial in human life, influencing perceptions, relationships, behaviour, and choices. Emotion recognition using Electroencephalography (EEG) in the Brain-computer Interface (BCI) domain presents significant challenges, particularly the need for extensive datasets. This study aims to generate synthetic EEG samples that are similar to real samples but are distinct by augmenting noise to a conditional denoising diffusion probabilistic model, thus addressing the prevalent issue of data scarcity in EEG research. The proposed method is tested on the DEAP dataset, showcasing upto 4.21% improvement in classification performance when using synthetic data. This is higher compared to the traditional GAN-based and DDPM-based approaches. The proposed diffusion-based approach for EEG data generation appears promising in refining the accuracy of emotion recognition systems and marks a notable contribution to EEG-based emotion recognition. Our research further evaluates the effectiveness of state-of-the-art classifiers on EEG data, employing both real and synthetic data with varying noise levels. © 2024, CC BY.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Sign Language Recognition using Deep Learning 15

Sign Language Recognition using Deep Learning

引用

15th International Conference on Developments in eSystems engineering, DeSE 2023

作者： Mahyoub, Mohamed Natalia, Friska Sudirman, Sud Mustafina, Jamila Tutor Reach United Kingdom Universitas Multimedia Nusantara Faculty of Engineering and Informatics Tangerang Indonesia School of Computer Science and Mathematics Liverpool John Moores University Liverpool United Kingdom Naberezhnye Chelny Institute Kazan Federal University Kazan Russia

ISBN: (纸本)9798350335149

Sign Language Recognition is a form of action recognition problem. The purpose of such a system is to automatically translate sign words from one language to another. While much work has been done in the SLR domain, it is a broad area of study and numerous areas still need research attention. The work that we present in this paper aims to investigate the suitability of deep learning approaches in recognizing and classifying words from video frames in different sign languages. We consider three sign languages, namely Indian Sign Language, American Sign Language, and Turkish Sign Language. Our methodology employs five different deep learning models with increasing complexities. They are a shallow four-layer Convolutional Neural Network, a basic VGG16 model, a VGG16 model with Attention Mechanism, a VGG16 model with Transformer Encoder and Gated Recurrent Units-based Decoder, and an Inflated 3D model with the same. We trained and tested the models to recognize and classify words from videos in three different sign language datasets. From our experiment, we found that the performance of the models relates quite closely to the model's complexity with the Inflated 3D model performing the best. Furthermore, we also found that all models find it more difficult to recognize words in the American Sign Language dataset than the others. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Benchmarking changepoint detection algorithms on cardiac time series

arXiv

引用

arXiv 2024年

作者： Cakmak, Ayse Selen Reinertsen, Erik Nemati, Shamim Clifford, Gari D. Department of Biomedical Informatics Emory University United States School of Electrical and Computer Engineering Georgia Institute of Technology United States Department of Biomedical Informatics University of California at San Diego United States

The pattern of state changes in a biomedical time series can be related to health or disease. This work presents a principled approach for selecting a changepoint detec- tion algorithm for a specific task, such as disease classification. Eight key algorithms were compared, and the performance of each algorithm was evaluated as a function of temporal tolerance, noise, and abnormal conduction (ectopy) on realistic artificial cardiovascular time series data. All algorithms were applied to real data (cardiac time series of 22 patients with REM-behavior disorder (RBD) and 15 healthy controls) using the parameters selected on artificial data. Finally, features were derived from the de- tected changepoints to classify RBD patients from healthy controls using a K-Nearest Neighbors approach. On artificial data, Modified Bayesian Changepoint Detection al- gorithm provided superior positive predictive value for state change identification while Recursive Mean Difference Maximization (RMDM) achieved the highest true positive rate. For the classification task, features derived from the RMDM algorithm provided the highest leave one out cross validated accuracy of 0.89 and true positive rate of 0.87. Automatically detected changepoints provide useful information about subject's phys- iological state which cannot be directly observed. However, the choice of change point detection algorithm depends on the nature of the underlying data and the downstream application, such as a classification task. This work represents the first time change point detection algorithms have been compared in a meaningful way and utilized in a classification task, which demonstrates the effect of changepoint algorithm choice on application *** Codes 92C30, 92C32, 03H10, 62H30, 68Q07, 8T07, 78-10, 92-10, 62R07, 68T09, 68T10 © 2024, CC BY-NC-SA.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Optimizing 3D Flight Paths for Multiple UAVs with Connectivity Management in Urban Delivery Systems 19

Optimizing 3D Flight Paths for Multiple UAVs with Connectivi...

引用

19th International Conference on Ubiquitous Information Management and Communication, IMCOM 2025

作者： Lee, Yeongmin Syahran, Raihan Muhammad Spangenberger, Mael Yang, Huigyu Oh, Jiyong Choo, Hyunseung Dept. of Artificial Intelligence Sungkyunkwan University Suwon Korea Republic of Dept. of Electrical and Computer Engineering Sungkyunkwan University Suwon Korea Republic of Convergence Research Institute Sungkyunkwan University Suwon Korea Republic of Dept. of AI Systems Engineering Sungkyunkwan University Suwon Korea Republic of Dept. of Informatics Télécom SudParis Évry France

ISBN: (纸本)9798331507817

The dynamic deployment of aerial vehicles in urban delivery scenarios demands precise route planning, reliable data links, and efficient use of network infrastructure. Although prior efforts have explored various aspects of unmanned navigation or communication optimization, existing approaches often overlook the combined impact of three-dimensional path selection and robust connectivity maintenance. To address this gap, this paper proposes a deep reinforcement learning framework employing Independent Proximal Policy Optimization (IPPO) to steer the flight paths of unmanned vehicles while curtailing needless interactions with base stations (BSs). By incorporating communication-specific metrics, Reference Signal Received Power (RSRP), the proposed system adapts to urban conditions and upholds resilient network links. Comprehensive performance evaluations reveal the potential of the framework to reduce excessive BS connections by up to 82.24%, preserve RSRP levels above -80 dBm, and balance handover frequencies across flight trajectories. These findings affirm the scalability and effectiveness of the method for achieving efficient aerial navigation and consistent communication in urban delivery services. © 2025 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

WildGaussians: 3D Gaussian splatting in the wild 24

WildGaussians: 3D Gaussian splatting in the wild

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Jonas Kulhanek Songyou Peng Zuzana Kukelova Marc Pollefeys Torsten Sattler Czech Institute of Informatics Robotics and Cybernetics Czech Technical University in Prague and Faculty of Electrical Engineering Czech Technical University in Prague and Department of Computer Science ETH Zurich Department of Computer Science ETH Zurich Visual Recognition Group Faculty of Electrical Engineering Czech Technical University in Prague Czech Institute of Informatics Robotics and Cybernetics Czech Technical University in Prague

ISBN: (纸本)9798331314385

While the field of 3D scene reconstruction is dominated by NeRFs due to their photorealistic quality, 3D Gaussian Splatting (3DGS) has recently emerged, offering similar quality with real-time rendering speeds. However, both methods primarily excel with well-controlled 3D scenes, while in-the-wild data - characterized by occlusions, dynamic objects, and varying illumination - remains challenging. NeRFs can adapt to such conditions easily through per-image embedding vectors, but 3DGS struggles due to its explicit representation and lack of shared parameters. To address this, we introduce WildGaussians, a novel approach to handle occlusions and appearance changes with 3DGS. By leveraging robust DINO features and integrating an appearance modeling module within 3DGS, our method achieves state-of-the-art results. We demonstrate that WildGaussians matches the real-time rendering speed of 3DGS while surpassing both 3DGS and NeRF baselines in handling in-the-wild data, all within a simple architectural framework.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimizing Light-Tree and Edge Resource Placement for Diverse Latency Needs

Optimizing Light-Tree and Edge Resource Placement for Divers...

引用

International Conference on Transparent Optical Networks

作者： P. Soumplis K. Yiannopoulos K. Christodoulopoulos E. Varvarigos Institute of Communication and Computer Systems Zografou Greece School of Electrical and Computer Engineering National Technical University of Athens Zografou Greece Department of Informatics and Telecommunications University of the Peloponnese Tripoli Greece Department of Informatics and Telecommunications National and Kapodistrian University of Athens Greece

ISBN: (数字)9798350377323

ISBN: (纸本)9798350377330

Coherent Point-to-MultiPoint (P2MP) transceivers are gaining attention for their efficacy in aggregating diverse workloads through optical light-trees, particularly in extended access scenarios where P2MP hub nodes integrate edge computing capabilities. This work addresses the challenge of optimizing the placement of both computing and networking resources to serve various workload quality classes, which can be suboptimal when treated independently. This is done taking into account the latency and network capacity requirements of the different classes, the optical connections physical layer performance, as well as computational costs.

关键词： Point-to-multipoint communications Integrated optics Network topology Computational modeling Optical fiber networks Physical layer Transceivers

来源：评论

学校读者我要写书评

暂无评论

Pending Interest Table (PIT) Performance Analysis in Named Data Networking on Palapa Ring Topology 16

Pending Interest Table (PIT) Performance Analysis in Named D...

引用

16th International Conference on Telecommunication Systems Services and Applications, TSSA 2022

作者： Sucipto, Adi Jupriyadi Ahdan, Syaiful Arifin, Hasan Nur Zaki Hamidi, Eki Ahmad Syambas, Nana Rachmana School of Electrical Engineering and Informatics Bandung Institute of Technology Bandung40132 Indonesia Universitas Teknokrat Indonesia Faculty of Engineering and Computer Science Lampung35132 Indonesia Uin Sunan Gunung Djati Department of Electrical Engineering Bandung40132 Indonesia

ISBN: (纸本)9781665455947

The concept of the internet in the future will prioritize content, by reducing delays in data transmission. Named Data Networking (NDN) is a content-based future internet concept that changes the paradigm of using IP. Inside the NDN router, there are three data structures, namely Content Store (CS), Pending Interest Table (PIT), and Forwarding Information Base (FIB). Pending Interest Table (PIT) contains a list of unfulfilled interests. This condition occurs when the node has not received a response after the interest forwarding process. Measurable and fast PIT performance is a challenge in Named Data Networks. In this study, we will try to do a simulation to measure and analyze the performance of PIT in NDN in the Palapa Ring topology. The research was conducted using the NDNSim simulator, to see the performance in the PIT. The simulation and analysis of the results show that the granularity of a prefix has an effect on In Satisfied Interest in an NDN network. At the number of interests of 100, the result obtained from the simulation is that there is a decrease in the percentage of interest data served, amounting to more than 20%. At the amount of interest in 1000 about more than 30%. The length of the prefix and the number of interest sent by the consumer affect the performance of the PIT, seen from the number of In Satisfied Interests. © 2022 IEEE.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Optical Evanescent Field for Direct Observation of Sub-100nm Particle Motion on Glass Surface 1

Optical Evanescent Field for Direct Observation of Sub-100nm...

引用

1st International Conference on Robotics, engineering, Science, and Technology, RESTCON 2024

作者： Khajornrungruang, Panart Permpatdechakul, Thitipat Suzuki, Keisuke Blattler, Aran Terayama, Yutaka Goto, Daiki Phaisalpanumas, Pipat Bun-Athuek, Natthaphon Phunopas, Amornphun Kyushu Institute of Technology Iizuka-Shi Mechanical Science and Technology Division Dept. of Intelligent and Control System Faculty of Computer Science and Systems Engineering Fukuoka Japan Kyushu Institute of Technology Iizuka-Shi Mechanical Information Science and Technology Division Dept. of Interdisciplinary Informatics Graduate School of Computer Science and Systems Engineering Fukuoka Japan King Mongkut's University of Technology North Bangkok Bangsue Faculty of Engineering Dept. of Production and Robotics Engineering Bangkok Thailand Kyushu Institute of Technology Iizuka-Shi Mechanical Science and Technology Course Dept. of Intelligent and Control System Graduate School of Computer Science and Systems Engineering Fukuoka Japan King Mongkut's University of Technology North Bangkok Bangsue Faculty of Engineering Dept. of Industiral Engineering Bangkok Thailand King Mongkut's University of Technology North Bangkok Bangsue College of Industrial Technology Dept. of Mechanical Engineering Technology Bangkok Thailand

ISBN: (纸本)9798350348842

To directly investigate the dynamic nanoscale phenomenon on the surface being processed in wet conditions such as precision polishing, and cleaning in semiconductor industrial, an optical method for visualization and observation of each sub-100 nm sized particle that is moving on an interface such as a silica glass surface by applying an evanescent field have been proposing. Subsequently, we developed an experimental apparatus equipped with an optical microscopy system for verifying the moving particle observation method in a laboratory scale. This article introduces some experimentally direct observation results of duplicated wet processes. © 2024 IEEE.

关键词： Visualization

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：