检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Tatsukichi Shibuya Nakamasa Inoue Rei Kawakami Ikuro Sato Department of Computer Science Institute of Science Tokyo Tokyo Japan Department of Systems and Control Engineering Institute of Science Tokyo Tokyo Japan Denso IT Laboratory Tokyo Japan

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

For deploying deep neural networks on edge devices with limited resources, binary neural networks (BNNs) have attracted significant attention, due to their computational and memory efficiency. However, once a neural network is binarized, finetuning it on edge devices becomes challenging because most conventional training algorithms for BNNs are designed for use on centralized servers and require storing real-valued parameters during training. To address this limitation, this paper introduces binary stochastic flip optimization (BinSFO), a novel training algorithm for BNNs. BinSFO employs a parameter update rule based on Boolean operations, eliminating the need to store real-valued parameters and thereby reducing memory requirements and computational overhead. In experiments, we demonstrated the effectiveness and memory efficiency of BinSFO in fine-tuning scenarios on six image classification datasets. BinSFO performed comparably to conventional training algorithms with a 70.7% smaller memory requirement. Code is released at https://***/TatsukichiShibuya/ICASSP2025_BinSFO

关键词： Training Speech coding Memory management Signal processing algorithms Signal processing Classification algorithms Servers Speech processing Optimization Image classification

来源：评论

学校读者我要写书评

暂无评论

A New Belief-Based Incomplete Pattern Unsupervised Classification Method : Extended Abstract 39

A New Belief-Based Incomplete Pattern Unsupervised Classific...

引用

39th IEEE International Conference on Data engineering, ICDE 2023

作者： Zhang, Zuowei Liu, Zhe Ma, Zongfang Zhang, Yiru Wang, Hao University of Rennes 1 Institut de Recherche en Informatique et Systèmes Aléatoires France Xi'an University of Architecture and Technology School of Information and Control Engineering China Norwegian University of Science and Technology Department of Computer Science Norway

ISBN: (纸本)9798350322279

Imputing the incomplete patterns in clustering tasks is a common but risky procedure, because the estimated values may affect the real distribution of the data and deteriorate the results. To address this problem, a new belief-based incomplete pattern unsupervised classification method (BPC) with uncertainty and imprecision reasoning is proposed in this paper. First, the complete patterns are grouped into a few clusters to obtain the corresponding reliable centers, and thereby are divided into reliable patterns and unreliable ones by an optimization method. Second, a basic classifier trained by reliable patterns is employed to classify unreliable patterns and incomplete patterns edited by the neighbors. Finally, some imprecise patterns are carefully reassigned again by a new distance-based rule depending on the obtained reliable centers and belief functions theory. The simulation results show that the BPC has the potential to deal with real datasets. © 2023 IEEE.

关键词： Uncertainty analysis

来源：评论

学校读者我要写书评

暂无评论

When Does Sora Show:The Beginning of TAO to Imaginative Intelligence and Scenarios engineering

引用

IEEE/CAA Journal of Automatica Sinica 2024年第4期11卷 809-815页

作者： Fei-Yue Wang Qinghai Miao Lingxi Li Qinghua Ni Xuan Li Juanjuan Li Lili Fan Yonglin Tian Qing-Long Han School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China State Key Laboratory for Management and Control of Complex Systems Chinese Academy of SciencesBeijing 100190China Faculty of Innovation Engineering Macao University of Science and TechnologyMacao 999078China Department of Electrical and Computer Engineering Indiana University-Purdue University IndianapolisIndianapolisIN 46202USA Faculty of Innovation Engineering Macao University of Science and TechnologyMacao 999078China Virtual Reality Fundamental Research Laboratory Department of Mathematics and TheoriesPeng Cheng LaboratoryShenzhen 518000China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China School of Information and Electronics Beijing Institute of TechnologyBeijing 100081China School of Science Computing and Engineering TechnologiesSwinburne University of TechnologyMelbourneVIC 3122Australia

DURING our discussion at workshops for writing“What Does ChatGPT Say:The DAO from Algorithmic Intelligence to Linguistic Intelligence”[1],we had expected the next milestone for Artificial Intelligence(AI)would be in the direction of Imaginative Intelligence(II),i.e.,something similar to automatic wordsto-videos generation or intelligent digital movies/theater technology that could be used for conducting new“Artificiofactual Experiments”[2]to replace conventional“Counterfactual Experiments”in scientific research and technical development for both natural and social studies[2]-[6].Now we have OpenAI’s Sora,so soon,but this is not the final,actually far away,and it is just the beginning.

关键词： something Intelligence replace

来源：评论

学校读者我要写书评

暂无评论

Experimental Evaluation of High-Precision System Clock Synchronization with BeiDou for Wide-Area Industrial Internet-of-Things 48

Experimental Evaluation of High-Precision System Clock Synch...

引用

48th Annual Conference of the IEEE Industrial Electronics Society, IECON 2022

作者： Yang, Fan Wang, Jinsong Ding, Yuemin Xing, Lantao Tianjin University of Technology School of Computer Science & Engineering Tianjin300384 China University of Navarra Tecnun School of Engineering San Sebastián20009 Spain Shandong University School of Control Science & Engineering Jinan250100 China

ISBN: (数字)9781665480253

ISBN: (纸本)9781665480253

High-precision clock synchronization is required for many industrial Internet of Things (IIoTs) supporting real-time monitoring and control for applications, such as the smart grid. For high-precision clock synchronization in IIoTs, several approaches can be used, such as the IEEE 1588 protocol, the Network Time Protocol (NTP), and the Global Navigation Satellite System (GNSS) based methods. Among them, the GNSS-based methods are more appropriate for IIoTs distributed in wide areas. It has been reported that nanosecond-level time synchronization can be achieved by GNSS receivers. However, the end-to-end accuracy of the system clocks in the application processor is a different story. In this study, an experimental testbed has been built using the open-source Linux system, the BeiDou receiver, and the open-source Raspberry Pi 4 hardware. The Pulse-Per-Second (PPS) signals from the BeiDou receiver are used as the reference to synchronize the system clocks. With a one-second synchronization interval, experiments indicate that the system clock achieves an accuracy of 1 μs with a success rate of 92.7% and 2 μs with a success rate of 99.7%. While increasing the synchronization interval, the system clock relies more on the local crystal oscillator, and the timing error increases to 100 μs in 500 seconds. © 2022 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Stock Price Forecasting Based on Feature Fusion Deepar Model

Stock Price Forecasting Based on Feature Fusion Deepar Model

引用

第35届中国控制与决策会议

作者： QingLin Xie Qi Lang Xiaodong Liu School of Control Engineering Dalian University of Technology School of Computer Science and Information Technology Northeast Normal University School of Control Science and Engineering Dalian University of Technology

Stock forecasting has always attracted the attention of the ***,existing studies rarely use long-term price patterns as input ***,this paper proposes a new deep learning model that integrates the features of longterm price models to predict stock closing *** clustering algorithm is used to extract the features of the long-term price model,and the prediction model is selected based on the deep autoregressive *** method of output probability distribution of this model is suitable for time series data with large uncertainty such as financial *** way of maximizing the likelihood function of the future sequence can better reflect the inherent randomness of the *** can not only predict the value,but also predict the future fluctuation,and has high prediction *** with other deep learning models,the results show that the feature fusion the deep autoregressive model has lower prediction error and higher goodness of fit.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Max Flow Vulnerability of Undirected Planar Networks 32

Max Flow Vulnerability of Undirected Planar Networks

引用

32nd Italian Symposium on Advanced Database Systems, SEBD 2024

作者： Balzotti, Lorenzo Franciosa, Paolo G. Department of Computer Control and Management Engineering "Antonio Ruberti" Sapienza University of Rome Via Ariosto 25 Rome00185 Italy Department of Statistical Science Sapienza University of Rome p.le Aldo Moro 5 Rome00185 Italy

In the real world, planar networks commonly model utility distribution networks such as gas pipelines, water distribution, city's electric grid, and transportation networks like railways and highways. Evaluating the vulnerability of these networks is crucial on several point of view: in a military context in order to prevent terrorist attack, or for maintenance to prevent wear and tear or natural disasters from inflicting extremely severe damage. This work aims to evaluate the vulnerability of undirected planar network studying the decrease of maximum st-flow when an edge or a vertex is removed, that is, we propose algorithms for computing the max flow vitality of edges and vertices. Our algorithms are applied to real world scenarios, and we prove that high vitality values are well approximated in time close to the time currently required to compute st-max flow O(n log log n). All our algorithms work in O(n) space. © 2024 Copyright for this paper by its authors.

关键词： Natural gas pipelines

来源：评论

学校读者我要写书评

暂无评论

End-to-end Flow Scheduling Optimization for Industrial 5G and TSN Integrated Networks

End-to-end Flow Scheduling Optimization for Industrial 5G an...

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Liu, Houling Li, Mingyan Gu, Fuqiang Li, Qihao Zhang, Weiting Guo, Songtao Chongqing University College of Computer Science Chongqing400044 China Ministry of Education of China Key Laboratory of System Control and Information Processing Shanghai China Jilin University College of Communication Engineering Changchun130015 China Beijing Jiaotong University School of Electronic and Information Engineering Beijing100091 China

ISBN: (纸本)9798350351255

The integrated of the fifth generation (5G) and time-sensitive networking (TSN) is a promising approach to meet the requirements of deterministic forwarding with extremely low latency and high flexibility for the Industrial Internet of Things (IIoT). However, due to the large dissimilarity of protocol stacks, the 5G system normally serves as a logical bridge for TSN, performing hold-and-forward operations in Base Stations (BSs) under the current 5G-TSN frameworks. To provide ubiquitous and seamless connectivity for IIoT devices, this paper focuses on the integrated enhancement of 5G and TSN by optimizing the scheduling of time-sensitive flows with end-to-end latency of 5G-TSN transmission taken into consideration. Specifically, we propose a novel architecture named GF-CQF, which combines 5G Grant-free (GF) access with TSN Cyclic Queuing and Forwarding (CQF). Subsequently, the distributed flow scheduling problem based on GF-CQF architecture is established due to the high timeliness demands. To alleviate the impact of the uncertainty inherent in 5G channels on end-to-end deterministic transmission, a feature-aware decentralized real-time scheduling (FDRS) policy based on Multi-agent Reinforcement Learning is proposed. FDRS allows each agent at BS to adaptively allocate TSN injection slots for flows mainly based on dynamic 5G transmission performance and TSN network resource state so that TSN queue overflow can be avoided and flow delay constraints can be guaranteed. Simulations show that FDRS offers superior scheduling capabilities under the limited TSN queue resources. © 2024 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

Multi-agent Deep Reinforcement Learning Based Multi-objective Charging control for Electric Vehicle Charging Station 7th

Multi-agent Deep Reinforcement Learning Based Multi-objectiv...

引用

7th Chinese Conference on Swarm Intelligence and Cooperative control, CCSICC 2023

作者： Gong, Jiabao Fu, Weiming Kang, Yu Qin, Jiahu Xiao, Feng Department of Automation University of Science and Technology of China Hefei230027 China State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources School of Control and Computer Engineering North China Electric Power University Beijing102206 China

ISBN: (纸本)9789819733279

This paper proposes a multi-agent deep reinforcement learning (MADRL) based algorithm for charging control of multiple electric vehicles (EVs) in an electric vehicle charging station (EVCS) with dynamic operations. By taking the uncertainties of EV users’ behaviors and charging prices into consideration, a Markov game with multiple objectives is first established, where the individual objective of reducing EV charging costs and the collective objective of mitigating charging load fluctuations are both studied. Then, a multi-agent actor-double-critic (MAADC) algorithm with a unified advantage actor-critic framework is proposed, where two advantage functions are designed as the local and global critics, and the integrated actor is employed to balance the learning towards the individual and collective objectives. The simulation results obtained in a delicately built EVCS environment show the effectiveness of MAADC. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Parallel Byzantine fault tolerance method for blockchain

Parallel Byzantine fault tolerance method for blockchain

引用

International Conference on Artificial Intelligence, Blockchain, Computing and Security, ICABCS 2023

作者： Dubey, Kumar Pradyot Gnanaprakasam, C.N. Khan, Ihtiram Raza Sadik, Md Shibli Al-Farhani, Liwa H. Ray, Samrat Department of Computer and Information Science Jagadguru Ram Bhadracharya Divyanga University UP Chitrakoot India Department of Electronics and Instrumentation Engineering St. Joseph’s College of Engineering Tamilnadu Chennai India Computer Science Department Jamia Hamdard Delhi India University of Portsmouth Portsmouth United Kingdom System Analysis Control and Information Processing Dep Academy of Engineering RUDN University Moscow Russia The Institute of Industrial Management Economics and Trade Peter The Great Saint Petersburg Polytechnic University Russia

ISBN: (纸本)9781032669663

Enterprise-focused blockchain applications that offer safe, anonymous, and immutable transactions are growing as blockchain technology matures. Traditional blockchain architecture has low performance and scalability, making it unsuitable for enterpriselevel applications with high concurrency and massive data. This paper studies and proposes a simplified Byzantine fault-tolerant (SBFT) consensus algorithm to improve consensus stage efficiency, a Task parallel smart contract model to maximise multi-core system parallel efficiency, and an improved blockchain system architecture to reflect the lightweight, low coupling nature of blockchain technology. It facilitates corporate application secondary development and the ParaChain blockchain and smart contract technology. Experimental verification compares the parallelization - based ParaChain blockchain’s TPS and scalability. PBFT-based blockchain systems have been greatly improved. © 2024 The Author(s).

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

An Isomerism Learning Model to Solve Time-Varying Problems Through Intelligent Collaboration

引用

IEEE/CAA Journal of Automatica Sinica 2023年第8期10卷 1772-1774页

作者： Zhihao Hao Guancheng Wang Bob Zhang Leyuan Fang Haisheng Li the Department of Computer and Information Science University of MacaoMacao 999078China the School of Data Science the Chinese University of Hong KongShenzhen 518172 Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen 518000 China Industrial Control Systems Cyber Emergency Response Team Beijing 100040China IEEE the College of Electrical and Information Engineering Hunan UniversityChangsha 410082 the Peng Cheng Laboratory Shenzhen 518000China the Beijing Key Laboratory of Big Data Technology for Food Safety Beijing Technology and Business UniversityBeijing 100048 the School of Computer Science and Engineering Beijing Technology and Business UniversityBeijing 100048China

Dear Editor,This letter deals with a solution for time-varying problems using an intelligent computational(IC)algorithm driven by a novel decentralized machine learning approach called isomerism *** order to meet the challenges of the model’s privacy and security brought by traditional centralized learning models,a private permissioned blockchain is utilized to decentralize the model in order to achieve an effective coordination,thereby ensuring the credibility of the overall model without exposing the specific parameters and solution process.

关键词： ism thereby letter

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：