检索结果-内蒙古大学图书馆

36th ieee international System-on-Chip conference (SOCC)

作者： Wu, Guan-Wei Chang, Cheng-Yang Wu, An-Yeu (Andy) Natl Taiwan Univ Dept Elect Engn Taipei Taiwan Natl Taiwan Univ Grad Inst Elect Engn Taipei Taiwan

ISBN: (纸本)9798350300116

Convolutional neural networks (CNNs) are leveraged in many applications, such as image classification and natural language processing (NLP) tasks. However, the hardware implementation of CNNs not only occupies a considerable amount of memory space but also incurs high computational demands, which can lead to significant energy cost for data transfer. Therefore, an emerging computing-in-memory (CIM)-based architecture has been proposed to alleviate the energy bottleneck of CNNs. Various model compression methods have recently been studied to enhance energy efficiency of the CIM-based accelerator, including quantization and pruning. However, these works are separately discussed with the static inference scenario. In this paper, the proposed De-C3 can dynamically and jointly design the compression strategy. Besides, it adopts trainable energy-aware thresholds for both quantization and pruning scenarios. experimental results based on the CIFAR-10 dataset show that our De-C3 achieves up to 3.4x the energy reduction compared to state-of-the-art works.

关键词： Convolutional neural network computing-in-memory energy-aware training dynamic inference

来源：评论

学校读者我要写书评

暂无评论

SgridNet: An efficient and Optimize Bagging Deep Learning for Intrusion Detection in Smart grid Networks 15

SGridNet: An Efficient and Optimize Bagging Deep Learning fo...

引用

15th international conference on computing Communication and Networking Technologies, ICCCNT 2024

作者： Sahu, Pranav Vyas, O.P. Vyas, Ranjana Iiit Allahabad Dept. of Information Technology Prayagraj India

ISBN: (纸本)9798350370249

We have developed a framework called SgridNet, a Bagging Neural Network model, for enhancing security in smart grid networks by detecting intrusions and classifying threats. This framework focuses on analyzing time-series TCP/IP packets to identify anomalies. SgridNet employs Bagging to aggregate predictions from multiple models trained on different subsets of the data, enhancing overall model robustness. This approach helps in effectively identifying new attack features and accurately classifying threats. Initial experiments demonstrate promising outcomes, with SgridNet significantly reducing model size. While this does not inherently lead to faster inference times due to the nature of bagging, it does improve computational efficiency and resource usage. evaluation on NSL-KDD datasets using KNN, SVM, and ensemble learning algorithms for binary and multiclass classification tasks shows competitive performance using metrics like Accuracy, Precision, Recall, and F-Measure. While KNN and SVM perform reasonably well across categories, ensemble learning excels in specific tasks. Comparison with a baseline model indicates SgridNet's superiority in terms of model size, computational efficiency, and marginal accuracy improvement. © 2024 ieee.

关键词： Cyber Physical Security Deep Learning Intrusion Detection Smart grid

来源：评论

学校读者我要写书评

暂无评论

NSDF-Services: Integrating Networking, Storage, and computing Services into a Testbed for Democratization of Data Delivery 16

NSDF-Services: Integrating Networking, Storage, and Computin...

引用

16th ieee/ACM international conference on Utility and Cloud computing (UCC) / 10th ieee/ACM international conference on Big Data computing, Applications and Technologies (BDCAT)

作者： Luettgau, Jakob Martinez, Heberth Olaya, Paula Scorzelli, Giorgio Tarcea, Glenn Lofstead, Jay Kirkpatrick, Christine R. Pascucci, Valerio Taufer, Michela Univ Tennessee Knoxville TN 37996 USA Univ Utah Salt Lake City UT 84112 USA Univ Michigan Ann Arbor MI USA Sandia Natl Labs Albuquerque NM USA Univ Calif San Diego La Jolla CA 92093 USA

ISBN: (纸本)9798400702341

The lack of a readily accessible, tightly integrated data fabric connecting high-speed networking, storage, and computing services remains a critical barrier to the democratization of scientific discovery. To address this challenge, we are building National science Data Fabric (NSDF), a holistic ecosystem to facilitate domain scientists in their daily research. NSDF comprises networking, storage, and computing services, as well as outreach initiatives. In this paper, we present a testbed integrating three services (i.e., networking, storage, and computing). We evaluate their performance. Specifically, we study the networking services and their throughput and latency with a focus on academic cloud providers;the storage services and their performance with a focus on data movement using file system mappers for both academic and commercial clouds;and computing orchestration services focusing on commercial cloud providers. We discuss NSDF's potential to increase scalability and usability as it decreases time-to-discovery across scientific domains.

关键词： High-performance computing Cloud computing Data democratization PerfSonar XRootD

来源：评论

学校读者我要写书评

暂无评论

Analysis of the Signal and computing Technology in Finite Space

Analysis of the Signal and Computing Technology in Finite Sp...

引用

2023 ieee international conference on electrical, Automation and Computer engineering, ICeACe 2023

作者： Zhao, Peng Guo, Kai Liu, Benyun Liu, Jinglong Tong, Xinyuan Li, Xinjie Yue, Caipeng Wang, Siwei Ma, Tan Shen, Tong Shandong Electric Power Research Institute State Grid Corporation of China Jinan250002 China State Grid Jinan Electric Power Company State Grid Corporation of China Jinan250002 China

ISBN: (纸本)9798350309614

The limited space health intelligence sensing program is developed, the main function is to coordinate the startup and work of each hardware module, and obtain and transmit the monitoring data of operators. By using fuzzy, neural network analysis, topology analysis and other models, the monitoring parameters are counted, analyzed and cross-verified to predict the risk level. This paper calculates, predicts and evaluates the limited space threshold, personnel or equipment moving track and location, intelligent perception terminal displays and evaluates the safety status of the job site, personnel, equipment, etc., in real time, to realize the rapid disposal of emergencies and the rapid implementation of emergency rescue. © 2023 ieee.

关键词： burn computer master control program programming programming language

来源：评论

学校读者我要写书评

暂无评论

Task Offloading and Resource Allocation in an RIS-Assisted NOMA-Based Vehicular edge computing

引用

ieee ACCeSS 2024年 12卷 124330-124348页

作者： Yakubu, Abdul-Baaki Abd el-Malek, Ahmed H. Abo-Zahhad, Mohammed Muta, Osamu elsabrouty, Maha M. Egypt Japan Univ Sci & Technol Dept Elect & Commun Engn New Borg El Arab City 21934 Alexandria Egypt Assiut Univ Fac Engn Dept Elect Engn Asyut 71515 Egypt Kyushu Univ Fac Informat Sci & Elect Engn Fukuoka 8120053 Japan

With the rise of intelligent transportation (ITS), autonomous cars, and on-the-road entertainment and computation, vehicular edge computing (VeC) has become a primary research topic in 6G and beyond communications. On the other hand, reconfigurable intelligent surfaces (RIS) are a major enabling technology that can help in the task offloading domain. This study introduces a novel VeC architecture that incorporates non-orthogonal multiple access (NOMA) and reconfigurable intelligent surfaces (RIS), where vehicles perform binary or partial computation offloading to edge nodes (eNs) for task execution. We construct a vehicle-to-infrastructure (V2I) transmission model by considering vehicular interference and formulating a joint task offloading and resource allocation (JTORA) problem with the goal of reducing total service latency and energy usage. Next, we decompose this problem into task offloading (TO) problem on the vehicle side and resource allocation (RA) problem on the eN side. Specifically, we describe offloading decisions and offloading ratios as a decentralized partially observable Markov decision process (Dec-POMDP). Subsequently, a multi-agent distributed distributional deep deterministic policy gradient (MAD4PG) is proposed to solve the TO problem, where every vehicular agent learns the global optimal policy and obtains individual decisions. Furthermore, a whale optimization algorithm (WOA) is used to optimize the phase shift coefficient of the RIS. Upon receiving offloading ratios and offloading decisions from vehicles, edge nodes utilize the Lagrange multiplier method (LMM) and Karush-Kuhn-Tucker (KKT) conditions to address the RA problem. Finally, we design a simulation model based on real-world vehicular movements. The numerical results demonstrate that, compared to previous algorithms, our proposed approach reduces the overall delay and energy consumption more effectively.

关键词： Reconfigurable intelligent surfaces Resource management Optimization NOMA edge computing Cloud computing energy consumption Deep reinforcement learning Reconfigurable intelligent surface non-orthogonal multiple access real-time task offloading vehicular edge computing multi-agent deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Neural Network-Based Approach to Forecasting Bankruptcy Using Feature Importance Analysis 3

A Neural Network-Based Approach to Forecasting Bankruptcy Us...

引用

3rd international conference on Sentiment Analysis and Deep Learning, ICSADL 2024

作者： Aatmaja, Aditi Bhagwat, Dhruv Prakash, B. Nallarasan, V. SRM Institute of Science and Technology Department of Computing Technologies Chennai India SRM Institute of Science and Technology Department of Networking and Communications Chennai India SRM Institute of Science and Technology School of Computing Department of Networking and Communications Chengalpattu Tamilnadu Kattankulathur India

ISBN: (纸本)9798350396157

This research work addresses the critical problem of developing a reliable bankruptcy prediction model. Accurately predicting bankruptcy is essential to the stability of the economy, and existing methods have limitations that may lead to business failure. In this research, we aim to overcome these limitations and develop a more accurate bankruptcy forecasting method. To do this, we discuss the issue of unequal class sizes and the relative importance of various forms of misclassification. After identifying the most important financial characteristics using feature importance analysis, we employ neural networks to predict if a company will fail in the next years. To improve the model's performance, we adjust the parameters using grid Search Cross Validation. The accuracy of bankruptcy predictions might be greatly increased with our suggested methodology, which would also increase economic stability. © 2024 ieee.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

experimental reservoir computing based on a laser subject to optoelectronic feedback

Experimental reservoir computing based on a laser subject to...

引用

2024 international conference Laser Optics, ICLO 2024

作者： Danilenko, G.O. Viktorov, e.A. Kovalev, A.V. ITMO University Institute of Advanced Data Transfer Systems Saint Petersburg Russia

We experimentally study a reservoir computing system based on a DFB laser diode with positive optoelectronic feedback. The system is operated in the vicinity of the Hopf bifurcation at 40 MHz input symbol rate. We eva... 详细信息

ISBN: (纸本)9798350390674

关键词： Hopf bifurcation

来源：评论

学校读者我要写书评

暂无评论

Research on enhancing Technology Transfer efficiency through Artificial Intelligence and Recommendation Algorithms 3

Research on Enhancing Technology Transfer Efficiency through...

引用

3rd international conference on Data Analytics, computing and Artificial Intelligence, ICDACAI 2024

作者： Fangyu, Gao Di, Chen Shuai, Yang Zhenghu, Hao Qiong, Wu Xu, Zhang China Electric Equipment Group Hunan University of Science and Technology Shanghai China Nankai University Shanghai China Shandong University Shanghai China Science and Technology Research Institute Tsinghua University Shanghai China Beijing State Grid Fuda Technology Development Co. Ltd. University of Melbourne Beijing China Beijing State Grid Fuda Technology Development Co. Ltd. University College London Beijing China

ISBN: (纸本)9798331519032

This paper adopts a language processing model and semantic recommendation model to realize the intelligent matching of scientific research achievements and market demand in the electrical equipment industry, aiming to improve the transformation process of scientific research and innovation achievements into practical applications. Through in-depth demand analysis, the key data characteristics of both market demand and scientific research achievement are defined, and the corresponding information feature database is constructed. Pre training based on Bert model can accurately understand the scientific and technological achievements, scientific and technological needs, scientific research literature, patent documents and other contents in the field of electrical equipment. On this basis, an LFM implicit semantic recommendation algorithm model is added to realize the accurate matching between scientific research achievements and market scientific and technological needs. It is the first time for this technology recommendation in the field of transformation of scientific and technological achievements of electrical equipment. © 2024 ieee.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Simulation and Intelligent Optimization Algorithm for Anti icing Coating Construction Process 3

Simulation and Intelligent Optimization Algorithm for Anti i...

引用

3rd international conference on Data Analytics, computing and Artificial Intelligence, ICDACAI 2024

作者： Jianwen, Sun Jing, Peng Zhanguo, Cao Yongjie, Nie Zeweiyi, Gong Electric Power Science Research Institute of Yunnan Power Grid Co. Ltd. Yunnan Kunming650217 China

ISBN: (纸本)9798331519032

In cold climate regions, excessive accumulation of ice and snow on various surfaces not only hinders the normal operation of infrastructure such as transportation and energy supply, but also poses a serious threat to human production and living activities, bringing huge economic losses and safety hazards. To effectively alleviate the negative impact of this natural phenomenon, anti icing coating materials have emerged. With their unique characteristics such as low surface energy or high lubricity, they significantly reduce the adhesion strength of ice on the surface of objects under low-temperature and severe cold weather conditions, thereby achieving effective suppression of icing phenomena. This article deeply explores the simulation and intelligent optimization algorithm research of anti icing coating construction technology, aiming to finely control and optimize the coating preparation and construction process through advanced theoretical analysis and high-precision numerical simulation technology, combined with intelligent algorithms such as particle swarm optimization (PSO) algorithm. The experimental results show that the optimization of the anti icing coating construction process through PSO not only significantly improves the uniformity and adhesion of the coating, but also significantly reduces the risk of icing of the coating under extreme weather conditions. © 2024 ieee.

关键词： Process control

来源：评论

学校读者我要写书评

暂无评论

energy-Aware Online Task Offloading and Resource Allocation for Mobile edge computing 43

Energy-Aware Online Task Offloading and Resource Allocation ...

引用

43rd ieee international conference on Distributed computing Systems (ICDCS)

作者： Liu, Yu Mao, Yingling Shang, Xiaojun Liu, Zhenhua Yang, Yuanyuan SUNY Stony Brook Dept Elect & Comp Engn Stony Brook NY 11794 USA SUNY Stony Brook Dept Appl Math & Stat Stony Brook NY 11794 USA

ISBN: (纸本)9798350339864

Mobile edge computing with the near-data processing paradigm can support applications requiring low latency and high computing capability, where energy cost is a significant part of the expenditure. This paper formulates and studies the problem of online joint task offloading and resource allocation for latency minimization subjecting to a time average energy cost constraint in mobile edge computing systems. The formulated problem has four time-variant system states, i.e., data lengths, task sizes, channel conditions, and electricity prices, which are modeled based on real-world data. At the beginning of each time slot, the system has to make five online decisions jointly: base station selection, server selection for task offloading, communication bandwidth allocation, computing resource allocation, and frequency scaling. We prove the offline version of the formulated problem is NP-hard. We design an online algorithm with a provable approximation ratio and low computational complexity for the proposed problem. In particular, it balances energy cost and latency based on the drift-plus-penalty algorithm and makes server and base station selection decisions using a game theoretic-based algorithm. We conduct extensive real-world data-driven simulations to evaluate the proposed algorithm. Simulation results show that the proposed approach outperforms popular baselines.

关键词： Online Task Offloading Mobile edge computing Frequency Scaling

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：