检索结果-内蒙古大学图书馆

6th International Conference on Electrical, control and Instrumentation engineering, ICECIE 2024

作者： Hirasawa, Kazuhiro Asada, Daiki Nagai, Shunya Oya, Hidetoshi Graduate School of Integrative Science and Engineering Tokyo City University Tokyo Japan Niihama College Department of Electronics and Control Engineering Ehime Japan Tokyo City University The Department of Computer Science Tokyo Japan

ISBN: (纸本)9798350380040

This paper proposes a new variable gain robust state observer for a class of uncertain nonlinear systems. The variable gain robust state observer proposed in this paper consists of fixed observer gain matrices and nonlinear modification functions which are determined by appropriate updating rules. It is shown that sufficient conditions for the existence of the proposed variable gain robust state observer can be reduced to solvability of Linear Matrix Inequalities (LMIs). Finally, we give a simple numerical example. © 2024 IEEE.

关键词： State estimation

来源：评论

学校读者我要写书评

暂无评论

Equalizer Zero-Determinant Strategy in Discounted Repeated Stackelberg Asymmetric Game

引用

Journal of Systems Science & Complexity 2024年第1期37卷 184-203页

作者： CHENG Zhaoyang CHEN Guanpu HONG Yiguang Key Laboratory of Systems and Control Academy of Mathematics and Systems ScienceChinese Academy of SciencesBeijing 100190China School of Mathematical Sciences University of Chinese Academy of SciencesBeijing 100049China School of Electrical Engineering and Computer Science KTH Royal Institute of Technology10044 StockholmSweden. Department of Control Science and Engineering Tongji UniversityShanghai 201804China Shanghai Research Institute for Intelligent Autonomous Systems Tongji UniversityShanghai 210201China

This paper focuses on the performance of equalizer zero-determinant(ZD)strategies in discounted repeated Stackelberg asymmetric *** the leader-follower adversarial scenario,the strong Stackelberg equilibrium(SSE)deriving from the opponents’best response(BR),is technically the optimal strategy for the ***,computing an SSE strategy may be difficult since it needs to solve a mixed-integer program and has exponential complexity in the number of *** this end,the authors propose an equalizer ZD strategy,which can unilaterally restrict the opponent’s expected *** authors first study the existence of an equalizer ZD strategy with one-to-one situations,and analyze an upper bound of its performance with the baseline SSE *** the authors turn to multi-player models,where there exists one player adopting an equalizer ZD *** authors give bounds of the weighted sum of opponents’s utilities,and compare it with the SSE ***,the authors give simulations on unmanned aerial vehicles(UAVs)and the moving target defense(MTD)to verify the effectiveness of the proposed approach.

关键词： Discounted repeated Stackelberg asymmetric game equalizer zero-determinant strategy strong Stackelberg equilibrium strategy

来源：评论

学校读者我要写书评

暂无评论

A Modified Dempster Shafer Approach to Classification in Surgical Skill Assessment 31

A Modified Dempster Shafer Approach to Classification in Sur...

引用

31st International Conference on Electrical engineering, ICEE 2023

作者： Iranfar, Arash Soleymannejad, Mohammad Moshiri, Behzad Taghirad, Hamid D. School of Electrical and Computer Engineering University College of Engineering University of Tehran Tehran Iran University of Waterloo Department of Electrical and Computer Engineering Waterloo Canada Industrial Control Center of Excellence Faculty of Electrical Engineering Tehran Iran

ISBN: (纸本)9798350312560

Artificial intelligence systems are usually implemented either using machine learning or expert systems. Machine learning methods are usually more accurate and applicable to a broader range of applications. Expert systems, on the other hand, require much less data for training and generate more comprehensible results. These characteristics are typically desired in the fields of surgery and medicine because there isn't much data available. In order to give a machine's decisions a deeper level of semantics, it is also advantageous to incorporate a doctor's expertise into it. Furthermore, it is safer to understand the reasoning behind a machine's choices. In this paper, a Dempster-Shafer Theory (DST) based expert system is suggested for the task of surgical training skill assessment. An interval-based probabilistic feature analysis was applied to the data to assign values to the mass functions. Zhang's rule of combination was applied to handle the conflicting evidence in the prediction phase. The performance of the proposed method was compared to another DST classifier, SVM, and XGBoost. Our method outperforms SVM and other DST classifiers, but it is not as precise as XGBoost. By reducing the size of the dataset, the added benefit of using an expert system as opposed to a machine learning method was explored further. The performance of the suggested method is not adversely affected by the size of the dataset, whereas the XGBoost classifier is. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

A Robust Guaranteed Cost Formation control for a Class of Uncertain Multi-Agent Systems for Event-Trigger-Based Variable Gain controllers 6

A Robust Guaranteed Cost Formation Control for a Class of Un...

引用

6th International Conference on Electrical, control and Instrumentation engineering, ICECIE 2024

作者： Asada, Daiki Oya, Hidetoshi Nagai, Shunya Hoshi, Yoshikatsu Graduate School of Integrative Science and Engineering Tokyo City University Tokyo Japan Tokyo City University The Department of Computer Science Tokyo Japan Niihama College Department of Electronics and Control Engineering Ehime Japan

ISBN: (纸本)9798350380040

In this paper, we propose a design method for a new event-trigger-based variable gain robust controller which achieves both consensus and guaranteed cost performance for a class of uncertain multi-agent systems (MASs) with the leader-follower structure. The proposed variable gain robust controller is composed of stabilizing state feedback laws, state feedback inputs for consensus, and compensation inputs with adjustable parameters, and is designed for giving consideration to relative positions explicitly. In this paper, we show that sufficient conditions for the existence of the proposed robust guaranteed cost formation controller are given in terms of Linear Matrix Inequalities (LMIs). Finally, a simple example is included. © 2024 IEEE.

关键词： State feedback

来源：评论

学校读者我要写书评

暂无评论

Hamiltonian learning using machine-learning models trained with continuous measurements

引用

Physical Review Applied 2024年第4期22卷 044080页

作者： Kris Tucker Amit Kiran Rege Conor Smith Claire Monteleoni Tameem Albash Department of Applied Mathematics Department of Computer Science Center for Quantum Information and Control Department of Electrical and Computer Engineering Paris France Center for Computing Research

We build upon recent work on the use of machine-learning models to estimate Hamiltonian parameters using continuous weak measurement of qubits as input. We consider two settings for the training of our model: (1) supervised learning, where the weak-measurement training record can be labeled with known Hamiltonian parameters, and (2) unsupervised learning, where no labels are available. The first has the advantage of not requiring an explicit representation of the quantum state, thus potentially scaling very favorably to a larger number of qubits. The second requires the implementation of a physical model to map the Hamiltonian parameters to a measurement record, which we implement using an integrator of the physical model with a recurrent neural network to provide a model-free correction at every time step to account for small effects not captured by the physical model. We test our construction on a system of two qubits and demonstrate accurate prediction of multiple physical parameters in both the supervised context and the unsupervised context. We demonstrate that the model benefits from larger training sets, establishing that it is “learning,” and we show robustness regarding errors in the assumed physical model by achieving accurate parameter estimation in the presence of unanticipated single-particle relaxation.

关键词： Entanglement detection Nonlocality Quantum parameter estimation Machine learning Weak measurements

来源：评论

学校读者我要写书评

暂无评论

Global practical stabilization of the double integrator system with an imperfect sensor and subject to a bounded disturbance

引用

control Theory and Technology 2024年第3期22卷 468-478页

作者： Zhiyuan Zhu Yuanlong Li Zongli Lin Department of Automation Shanghai Jiao Tong Universityand Key Laboratory of System Control and Information Processing of Ministry of EducationShanghai 200240China Charles L.Brown Department of Electrical and Computer Engineering University of VirginiaP.0.Box 400743CharlottesvilleVA 22904-4743USA

This paper is concerned with global practical stabilization of the double integrator system with an imperfect sensor and subject to an additive bounded output *** imperfect sensor nonlinearity possesses the nonlinear characteristics of saturation and dead *** of the presence of output dead zone and the additive disturbance,the states cannot be expected to driven into an arbitrarily small neighborhood of the *** solve the global practical stabilization problem,we proposes a low gain-based linear dynamic output feedback law,under which the first state enters and remains in a bounded set whose size is depended on the bound of disturbance and the range of dead zone and the second state enters and remains in a pre-specified arbitrarily small set,both in finite *** results illustrate the effectiveness of our proposed control method.

关键词： Output saturation Disturbance rejection Output dead zone Low gain feedback-Linear output feedback Practical stabilization

来源：评论

学校读者我要写书评

暂无评论

Optimizing DNN Inference on Multi-Accelerator SoCs at Training-time

引用

IEEE Transactions on computer-Aided Design of Integrated Circuits and Systems 2025年

作者： Risso, Matteo Burrello, Alessio Pagliari, Daniele Jahier Politecnico di Torino Department of Control and Computer Engineering Turin10129 Italy

The demand for executing Deep Neural Networks (DNNs) with low latency and minimal power consumption at the edge has led to the development of advanced heterogeneous Systems-on-Chips (SoCs) that incorporate multiple specialized computing units (CUs), such as accelerators. Offloading DNN computations to a specific CU from the available set often exposes accuracy vs efficiency trade-offs, due to differences in their supported operations (e.g., standard vs. depthwise convolution) or data representations (e.g., more/less aggressively quantized). A challenging yet unresolved issue is how to map a DNN onto these multi-CU systems to maximally exploit the parallelization possibilities while taking accuracy into account. To address this problem, we present ODiMO, a hardware-aware tool that efficiently explores fine-grain mapping of DNNs among various on-chip CUs, during the training phase. ODiMO strategically splits individual layers of the neural network and executes them in parallel on the multiple available CUs, aiming to balance the total inference energy consumption or latency with the resulting accuracy, impacted by the unique features of the different hardware units. We test our approach on CIFAR-10, CIFAR-100, and ImageNet, targeting two open-source heterogeneous SoCs, i.e., DIANA and Darkside. We obtain a rich collection of Paretooptimal networks in the accuracy vs. energy or latency space. We show that ODiMO reduces the latency of a DNN executed on the Darkside SoC by up to 8× at iso-accuracy, compared to manual heuristic mappings. When targeting energy, on the same SoC, ODiMO produced up to 50.8× more efficient mappings, with minimal accuracy drop ( © 1982-2012 IEEE.

关键词： System-on-chip

来源：评论

学校读者我要写书评

暂无评论

The Dark Side of the Language: Pre-trained Transformers in the DarkNet

The Dark Side of the Language: Pre-trained Transformers in t...

引用

2023 International Conference Recent Advances in Natural Language Processing: Large Language Models for Natural Language Processing, RANLP 2023

作者： Ranaldi, Leonardo Nourbakhsh, Aria Patrizi, Arianna Ruzzetti, Elena Sofia Onorati, Dario Mastromattei, Michele Fallucchi, Francesca Zanzotto, Fabio Massimo Human-Centric ART Group Department of Enterprise Engineering University of Rome Tor Vergata Italy Department of Computer Control and Management Engineering University of Rome Sapienza Italy

ISBN: (纸本)9789544520922

Pre-trained Transformers are challenging human performances in many NLP tasks. The massive datasets used for pre-training seem to be the key to their success on existing tasks. In this paper, we explore how a range of pretrained Natural Language Understanding models perform on definitely unseen sentences provided by classification tasks over a DarkNet corpus. Surprisingly, results show that syntactic and lexical neural networks perform on par with pre-trained Transformers even after fine-tuning. Only after what we call extreme domain adaptation, that is, retraining with the masked language model task on all the novel corpus, pre-trained Transformers reach their standard high results. This suggests that huge pre-training corpora may give Transformers unexpected help since they are exposed to many of the possible sentences. © 2023 Incoma Ltd. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimization of Neural Networks using Swarm Intelligence Techniques for Achieving Energy Efficiency in Smart Building Architecture 1

Optimization of Neural Networks using Swarm Intelligence Tec...

引用

1st IEEE International Conference on Cognitive Robotics and Intelligent Systems, ICC - ROBINS 2024

作者： Raja, M. Karthick Shermatova, Makhbuba Al Ansari, Mohammed Saleh Abdufattokhov, Shokhjakhon Rao, Vuda Sreenivasa Raj, I. Infant Sri Eshwar College of Engineering Department of Computer Science and Engineering Coimbatore India Turin Polytechnic University in Tashkent Civil Engineering and Architecture Department Tashkent Uzbekistan University of Bahrain College of Engineering Department of Chemical Engineering Bahrain Turin Polytechnic University Automatic Control and Computer Engineering Dept Italy Tashkent Intl. University of Education Dept of Information Technologies Tashkent Uzbekistan Koneru Lakshmaiah Education Fou. Green Fields Department of Computer Science and Engineering A.P Vaddeswaram522302 India K. Ramakrishnan College of Engineering Department of Computer Science and Engineering Tamil Nadu Trichy621 112 India

ISBN: (纸本)9798350372748

The increasing prevalence of smart building architectures, driven by the integration of Internet of Things (IoT) devices and automation systems, has led to a surge in energy consumption. This research explores the application of swarm intelligence techniques as an innovative approach to optimize neural networks, aiming to strike a balance between maintaining the desired performance levels and minimizing energy consumption. The study investigates the integration of swarm-based optimization algorithms, such as Particle Swarm Optimization (PSO) into the training and operation of neural networks. These algorithms enable the networks to dynamically adapt and optimize their parameters in response to changing environmental conditions and user requirements. The research focuses on developing a comprehensive framework that considers the specific challenges posed by smart building architectures, including real-time data processing, sensor integration, and adaptive control. The proposed approach aims to achieve optimal neural network configurations that minimize energy consumption while ensuring reliable and responsive operation of smart building systems. The results demonstrate the potential of swarm intelligence to significantly improve the energy efficiency of neural network-enabled smart building architectures, providing a promising avenue for sustainable and intelligent infrastructure. The proposed model has an accuracy of 98.23% which is 7.64% higher than that of the traditional approaches. © 2024 IEEE.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

An iterative algorithm for phase hologram generation with overlap constraints 8

An iterative algorithm for phase hologram generation with ov...

引用

8th International Conference on Speckle Metrology, Speckle 2023

作者： Lin, Angyi Chen, Rui Lin, Huixiang You, Jiangfeng Zhang, Fucai The Department of Electrical and Electronic Engineering Southern University of Science and Technology Shenzhen518055 China School of Control and Computer Engineering North China Electric Power University Baoding071003 China

ISBN: (数字)9781510674592

ISBN: (纸本)9781510674585

Currently, the dominant encoding method for computer-generated holograms is phase-only holography. The Gerchberg-Saxton (GS) algorithm is a commonly used approach for hologram generation. However, it suffers from issues such as poor convergence and low efficiency. In this paper, we introduce an iterative hologram algorithm that addresses these challenges. Our proposed algorithm divides the target object into segments with a certain degree of data redundancy. It computes the hologram iteratively, giving extra weight to each iteration. We also incorporate overlap constraints to enhance hologram quality and accelerate convergence by leveraging data redundancy. What sets our algorithm apart from others is that it varies the order in which segments are processed during iterations, resulting in different final holograms. By randomly selecting the order in each iteration, we can rapidly generate multiple diverse holograms. Importantly, objects reconstructed from these holograms exhibit consistency, allowing for the quick generation of multiple phase-only holograms that help mitigate the speckle effect. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： Holograms

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：