检索结果-内蒙古大学图书馆

International Conference on Information Networking

作者： Edenilson Jônatas Dos Passos Adriano Fiorese Department of Computer Science (DCC) Graduate Program in Apllied Computing (PPGCAP) Santa Catarina State University (UDESC) Joinville Brazil

ISBN: (数字)9798331506940

ISBN: (纸本)9798331506957

Recent advancements in on-demand multimedia streaming have revealed the potential of this business model, driven by the convenience of accessing content anytime and anywhere. However, delivering a seamless user experience remains challenging due to the demanding nature of audiovisual streams, despite significant resource allocation. A promising solution involves using load balancers to distribute workloads evenly among processing units. This study proposes optimizing the processing and transmission resources by means of Software-Defined Networks (SDN). By continuously monitoring performance metrics like CPU usage and throughput, the SDN network controller determines the best server for handling new connections, even allowing seamless server migration during playback. The results show up to 80% faster response times and more consistent video quality, with reduced latency and uninterrupted playback. The approach also confirms the feasibility of SDN in streaming services, providing a foundation for future network improvements.

关键词： Transform coding Streaming media Load management User experience Quality assessment Servers Time factors Software defined networking Video recording Monitoring

来源：评论

学校读者我要写书评

暂无评论

FIFS: A Machine Learning-Based Fast AV1 Interpolation Filter Search

FIFS: A Machine Learning-Based Fast AV1 Interpolation Filter...

引用

IEEE Latin American Symposium on Circuits and Systems (LASCAS)

作者： William Kolodziejski Marcelo Porto Luciano Agostini Video Technology Research Group (ViTech) Graduate Program in Computing (PPGC) Federal University of Pelotas (UFPel) Pelotas RS Brazil

ISBN: (数字)9798331522124

ISBN: (纸本)9798331522131

AV1 is a codec developed by huge technology companies to be used in current and future commercial video applications. It introduces and improves several tools from its predecessor VP9, designed for various video scenarios. One of the improved tools is the Fractional Motion Estimation (FME), which generates sub-pixel predictors. AV1 employs four sets of interpolation filters, requiring significant computational effort during the Interpolation Filter Search (IFS) to identify the best filter to be used. This work proposes the FIFS, a machinelearning method developed to reduce the processing time of IFS while maintaining minimal impact on coding efficiency. This method achieves over 52% reductions in IFS time, with only a slight increase in BD-BR of $\mathbf{0.14 \%}$ . To the best of the authors' knowledge, this is the first work in the literature to propose a machine learning-based approach for the AV1 IFS.

关键词： Video coding Interpolation Filters Codecs Motion estimation Computational modeling Focusing Machine learning Companies Encoding

来源：评论

学校读者我要写书评

暂无评论

Compressing VAE-Based Out-of-Distribution Detectors for Embedded Deployment

arXiv

引用

arXiv 2024年

作者： Bansal, Aditya Yuhas, Michael Easwaran, Arvind College of Computing and Data Science Energy Research Institute @ NTU Interdisciplinary Graduate Program Nanyang Technological University Singapore

Out-of-distribution (OOD) detectors can act as safety monitors in embedded cyber-physical systems by identifying samples outside a machine learning model’s training distribution to prevent potentially unsafe actions. However, OOD detectors are often implemented using deep neural networks, which makes it difficult to meet real-time deadlines on embedded systems with memory and power constraints. We consider the class of variational autoencoder (VAE) based OOD detectors where OOD detection is performed in latent space, and apply quantization, pruning, and knowledge distillation. These techniques have been explored for other deep models, but no work has considered their combined effect on latent space OOD detection. While these techniques increase the VAE’s test loss, this does not correspond to a proportional decrease in OOD detection performance and we leverage this to develop lean OOD detectors capable of real-time inference on embedded CPUs and GPUs. We propose a design methodology that combines all three compression techniques and yields a significant decrease in memory and execution time while maintaining AUROC for a given OOD detector. We demonstrate this methodology with two existing OOD detectors on a Jetson Nano and reduce GPU and CPU inference time by 20% and 28% respectively while keeping AUROC within 5% of the baseline. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Integrity of Cyber-Physical Discrete Event Systems under covert actuator attacks

引用

IFAC-PapersOnLine 2024年第1期58卷 198-203页

作者： Samuel Oliveira André B. Leal Marcelo Teixeira Yuri K. Lopes Department of Exact and Technological Sciences Federal University of Amapá (UNIFAP) Macapá/AP Brazil Graduate Program in Electrical Engineering Santa Catarina State University (UDESC) Joinville/SC Brazil Graduate Program in Electrical Engineering Federal University of Technology (UTFPR) Pato Branco/PR Brazil Graduate Program in Applied Computing Santa Catarina State University (UDESC) Joinville/SC Brazil

The connectivity requirements brought by Industry 4.0 pose new challenges for cyber-physical systems (CPS), such as the strengthening of their resilience against cyberattacks. In this paper, we use the discrete event systems (DES) formalism to develop an approach that aims to protect CPS from covert actuator attacks using event permutation maps. In this approach, properly selected controllable events are permuted on the plant's site so that attacks on actuators do not have the impact on the plant as expected by the intruder and unsafe states are avoided. Furthermore, a property termed AE-protectability is presented to characterize the necessary and sufficient conditions for the system to be protected against such attacks through event permutations. The proposed approach is applied to a case study where the Identification of the attack is done without causing damage while maintaining the integrity of the system.

关键词： Cyber-security Supervisory Control Theory Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Neural Network Based Light Field Predictors: An Evaluation

Neural Network Based Light Field Predictors: An Evaluation

引用

IEEE Latin American Symposium on Circuits and Systems (LASCAS)

作者： Matheus Jahnke Ítalo D. Machado Lucas Ikenoue Bruno Zatt Video Technology Research Group (ViTech) Graduate Program in Computing (PPGC) Federal University of Pelotas (UFPel) Pelotas Brazil

Processing and storing the 4D structure of light fields can be challenging and expensive due to the high-dimensional data and its unique characteristics. There are plenty of works employing convolutional neural networks (CNNs) for light field prediction and encoding. Nonetheless, to the best of our knowledge, the literature lacks an efficiency evaluation of different CNN architectures as well as 4D neural networks for these purposes. Therefore, this paper presents an experimental study that assesses the performance of pipeline and U-net convolutional neural networks for light field block prediction in both spatial and angular dimensions. Additionally, we compare these architectures with a novel 4D network that aims to exploit the light field data structure. The results of the study show that U-net and 4D networks outperform classical CNN architectures in terms of prediction accuracy and residue generation. Furthermore, the spatial dimension prediction provides more valuable information for the networks to learn, improving their prediction by 5dB.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-Codec Video Quality Enhancement Model Based on Spatio- Temporal Deformable Fusion

Multi-Codec Video Quality Enhancement Model Based on Spatio-...

引用

IEEE Latin American Symposium on Circuits and Systems (LASCAS)

作者： Gilberto Kreisler Garibaldi Da Silveira Bruno Zatt Daniel Palomino Guilherme Corrêa Video Technology Research Group (ViTech) Graduate Program in Computing (PPGC) Federal University of Pelotas (UFPel) Pelotas Brazil

The popularization of mobile phones and other multimedia portable devices paved the way for the increase in video consumption worldwide. However, it is impossible to transmit a non-compressed video due to the high bandwidth required. To achieve significant compression rates, video codecs usually employ methods that damage the visual quality perceived by the end user in non-negligible levels. Different architectures based on deep learning have been recently proposed for Video Quality Enhancement (VQE). Still, most of them are trained and validated using videos generated by a single codec under fixed configurations. With the increase of video coding formats and standards on the market, VQE methods that apply to different contexts are desired. This paper proposes a new VQE model based on the Spatio- Temporal Deformable Fusion (STDF) archi-tecture, providing quality gains for videos compressed according to different formats and standards, such as HEVC, VVC, VP9, and AVI. The results demonstrate that by considering different video coding standards and formats to build the STDF model, a significant increase in VQE is achieved, with an average PSNR increment of up to 0.382 dB.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learning from failure - Development of a Culture of Failure Scale for Higher Education

Learning from failure - Development of a Culture of Failure ...

引用

2022 IEEE IFEES World Engineering Education Forum - Global Engineering Deans Council, WEEF-GEDC 2022

作者： Malschutzky, Michael M. Gurganus, Jamie R. Department of Management Sciences Germany Mechanical Engineering and Graduate School University of Maryland Baltimore County Engineering and Computing Education Program Baltimore United States

ISBN: (纸本)9781665475280

Benefits of a failure friendly culture, e.g., learning from failure, are widely known in occupational settings. Validated scales have been developed to measure organizational failure culture and individuals' mind-sets on failure. While research on learning from failure exists in secondary education, scales for higher education are lacking. To support both higher education engineering instructors and students, this contribution aims to develop a Culture of Failure Scale to assess students' and lecturers' handling of failures in engineering higher education settings. This data-driven approach helps inform instructors about aspects of educational settings that are safe and conducive to learning and supports their students to develop a mind-set where failures are seen as a source for improvements and learning. Exemplary application of the developed scale is shown by assessing the failure culture in different groupings, e.g., gender, between two participating German and US institutions. © 2022 IEEE.

关键词： Failure (mechanical)

来源：评论

学校读者我要写书评

暂无评论

HistAttentionNAS: A CNN built via NAS for Penile Cancer Diagnosis using Histopathological Images 16

HistAttentionNAS: A CNN built via NAS for Penile Cancer Diag...

引用

16th International Conference on ENTERprise Information Systems, CENTERIS 2024 - 12th International Conference on Project MANagement, ProjMAN 2024 - 14th International Conference on Health and Social Care Information Systems and Technologies, HCist 2024

作者： Durand R, José Ribamar Junior, Geraldo Braz Da Silva, Italo Francyles Santos Da Costa Oliveira, Rui Miguel Gil Applied Computing Group Federal University of Maranhão Vila Bacanga São Luís65080-805 Brazil Graduate Program in Adult Health - PPGSAD Federal University of Maranhão Vila Bacanga São Luís65080-805 Brazil

Penile cancer, although rare, has an increasing mortality rate in Brazil, highlighting the need for effective diagnostic methods. Artificial Intelligence (AI) in histopathological analysis can speed up and objectify diagnosis, but designing an ideal architecture is challenging. In this study, we propose a neural architecture search (NAS) methodology for detecting penile cancer in digital histopathology images. We explored different configurations of stem blocks and the inclusion of attention mechanisms, highlighting specific preferences depending on the magnification of the images. The results showed that the NAS methodology enabled the discovery of more accurate and optimized architectures for this task, surpassing conventional models. The proposed models achieve 89.5% and 88.5% F1-Score for 40X and 100X magnification, respectively. © 2025 The Author(s).

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Optimizing in-band network telemetry problem

引用

Procedia Computer Science 2024年 239卷 2074-2081页

作者： Thierno Bocar N’Diaye Marcelo Caggiani Luizelli Jules Degila Luciana Salete Buriol Institut de Mathematiques et Sciences Physiques Universite d’Abomey-Calavi (UAC) Benin Graduate Program in Electrical Engineering Universidade Federal do Pampa (UNIPAMPA) Brazil Graduate Program in Computing Universidade Federal do Rio Grande do Sul (UFRGS) Brazil

In-band network telemetry is a powerful framework for network monitoring. It allows the collection of telemetry data in real-time and provides network-wide visibility. However, depending on the routing of network flows and which telemetry data are collected, the network-wide visibility and the performance of monitoring applications may decrease. In this paper, we present the in-band network telemetry problem and extend the existing mathematical optimization models of the problem by proposing a new model that computes the routing of network flows. Results show that the new model outperforms existing models in term of network coverage and monitoring applications performance. The results of this work can be useful for network managers and enterprises to gain real-time insights into network performance.

关键词： Network monitoring Monitoring application Optimization In-band network telemetry Network flows routes

来源：评论

学校读者我要写书评

暂无评论

Use of Language to Generate Architectural Scenery with AI-Powered Tools 20th

Use of Language to Generate Architectural Scenery with AI-Po...

引用

20th International Conference on Computer-Aided Architectural Design Futures, CAAD Futures 2023

作者： Karahan, Hanım Gülsüm Aktaş, Begüm Bingöl, Cemal Koray Architectural Design Computing Graduate Program Istanbul Technical University Istanbul34367 Turkey Department of Architecture Altınbaş University Istanbul Turkey Department of Management in the Built Environment Delft University of Technology Delft Netherlands

ISBN: (纸本)9783031371882

The quality of communication with a computer impacts how the designer performs during the design process. Today, Artificial Intelligence (AI) empowers the designer by expanding the solution space using the expertise from previous knowledge. However, the developments in AI-powered design tools mainly focus on visual and spatial enhancements. In the last decade, AI-powered design tools mostly experimented with image transformation models (GANs) to provide fast insights to designers using learned experiences, simulations, or datasets. The studies on the design process using verbal language with the help of AI are limited. Therefore, designers’ capacity to communicate with intelligent machines would lead us to envision the future of AI-powered design tools. In design practice, designers develop individual and contextual studies through digital tools. This study investigates the process of architectural visual generation and verbal communication to describe architectural images by architecture graduates with prior experience or no experience in prior with Midjourney. The research focuses on the designers’ semantic language during the design process with the AI-powered tool and analysis of the verbal part of the communication. The results of this study show that participants’ first impressions of the image and how they express their impressions through description do not correspond with how Midjourney interprets those descriptions. Furthermore, architects’ image generation process using the tool is nonlinear. As architects develop a deeper understanding of changing modes of interactions, they are more likely to benefit from AI-powered tools as collaborative entities. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Artificial intelligence

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：