检索结果-内蒙古大学图书馆

Research Square 2023年

作者： Gatto, Elaine Cecília Ferrandin, Mauri Cerri, Ricardo Department of Computer Science Federal University of São Carlos SP São Carlos13565-905 Brazil Department of Control Automation and Computing Engineering Federal University of Santa Catarina SC Blumenau89036-002 Brazil

Multi-Label Classification is the task of simultaneously predicting a set of labels for an instance. Typically, two approaches are used: global, which trains a single classifier to deal with all classes at once, and local, which divides the problem into many binary problems. In both approaches, learning label correlations is still an open issue. In this paper, we propose a method to cluster the label space and find partitions of disjoint correlated labels called hybrid partition, which can be considered in-between the local and global, combining the benefits of both. Our proposal entails: i) clustering the label space based on label correlations;ii) generating and validating the resulting hybrid partitions;iii) selecting the optimal partitions between them;and iv) evaluating their performances. Since we are also interested in analyzing how hybrid partitions are close to the best possible partitions, we compared our results with the ones obtained using partitions found by an oracle, an exhaustive, and a random search. The oracle and exhaustive strategies generate all possible label partitions for a dataset. While the oracle chooses the best one based on the test set alone, the exhaustive choices are based on validation. Experiments on eight well-known multi-label datasets revealed that both our proposal and the random partitions achieve superior or competitive results compared with the others. This indicates that the conventional global and local approaches may not accurately account for label correlations, requiring significant improvements. In this case, clustering the label space can be a solution. © 2023, CC BY.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

SymAware: A Software Development Framework for Trustworthy Multi-Agent Systems with Situational Awareness

arXiv

引用

arXiv 2024年

作者： Casablanca, Ernesto Zhang, Zengjie Marchesini, Gregorio Haesaert, Sofie Dimarogonas, Dimos V. Soudjani, Sadegh School of Computing Newcastle University Newcastle United Kingdom Department of Electrical Engineering Eindhoven University of Technology Eindhoven Netherlands Division of Decision and Control Systems KTH Royal Institute of Technology Stockholm Sweden Max Planck Institute for software systems Germany

Developing trustworthy multi-agent systems for practical applications is challenging due to the complicated communication of situational awareness (SA) among agents. This paper showcases a novel efficient and easy-to-use software framework for multi-agent simulation, named SymAware which provides a rich set of predefined data structures to compute, store, and communicate SA for agents. It also provides an abstract interface for the agents to compute their control inputs taking into account the awareness of the situation, knowledge, and risk of surrounding agents. Besides, utilizing a cluster of specialized components, SymAware hides the heavy computation of physical rendering and communication interfacing of simulation engines behind the control threads, resulting in high implementation efficiency in bridging the gap between conceptual prototyping and practical applications. Three multiagent case studies are used to validate the efficacy and efficiency of this software framework. © 2024, CC BY.

关键词： Software design

来源：评论

学校读者我要写书评

暂无评论

Analytical-Chemistry-Informed Transformer for Infrared Spectra Modeling 39

Analytical-Chemistry-Informed Transformer for Infrared Spect...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Huang, Shiluo Jin, Yining Jin, Wei Mu, Ying School of Computing and Artificial Intelligence Southwestern University of Finance and Economics Chengdu China Department of Electrical and Computer Engineering University of Alberta Edmonton AB Canada Research Centre for Analytical Instrumentation State Key Laboratory of Industrial Control Technology Zhejiang University Hangzhou China

ISBN: (纸本)157735897X

Infrared (IR) spectroscopy is a fundamental technique in analytical chemistry. Recently, deep learning (DL) has drawn great interest as the modeling method of infrared spectral data. However, unlike vision or language tasks, IR spectral data modeling is faced with the problem of calibration transfer and has distinctive characteristics. Introducing the prior knowledge of IR spectroscopy could guide the DL methods to learn representations aligned with the domain-invariant characteristics of spectra, and thus improve the performance. Despite such potential, there is a notable absence of DL methods that incorporate such inductive bias. To this end, we propose Analytical-Chemistry-Informed Transformer (ACT) with two modules informed by the field knowledge in analytical chemistry. First, ACT includes learnable spectral processing inspired by chemometrics, which comprises spectral pre-processing, tokenization, and post-processing. Second, a straightforward yet effective representation learning mechanism, namely spectral-attention, is incorporated into ACT. Spectral-attention utilizes the intra-spectral and inter-spectral correlations to extract spectral representations. Empirical results show that ACT has achieved competitive results in 9 analytical tasks covering applications across pharmacy, chemistry, and agriculture. Compared with existing networks, ACT reduces the root mean square error of prediction (RMSEP) by more than 20% in calibration transfer tasks. These results indicate that DL methods in IR spectroscopy could benefit from the integration of prior knowledge in analytical chemistry. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Unlocking Freeform Structured Surface Denoising with Small Sample Learning: Enhancing Performance Via Physics-Informed Loss and Detail-Driven Data Augmentation

SSRN

引用

SSRN 2024年

作者： Cui, Weixin Lou, Shan Zeng, Wenhan Kadirkamanathan, Visakan Qin, Yuchu Scott, Paul J. Jiang, Xiangqian EPSRC Future Advanced Metrology Hub School of Computing and Engineering University of Huddersfield HuddersfieldHD1 3DH United Kingdom Department of Automatic Control and Systems Engineering The University of Sheffield Mappin Street SheffieldS1 3JD United Kingdom

Denoising plays a vital role in surface metrology and other engineering domains. Frequently, these noises manifest themselves as outliers and high-frequency artifacts. Traditional denoising techniques, such as Gaussian and partial differential equation-based diffusion filters, although effective, often involve a time-consuming calibration process, especially when addressing diverse topographical surfaces. The challenge intensifies for novice users in the presence of varied surface textures. This paper addresses these challenges by introducing a learning-based solution that aims to deliver the benefits of traditional filters without the need for manual tuning. Denoising convolutional neural network is employed as the basic architecture due to its effectiveness in tackling mix-level Gaussian noise and adapting to small training dataset. Acknowledging the constraints of limited datasets, prevalent in many engineering scenarios, we propose a novel physics-informed loss. This innovative loss marries filtering techniques with neural network training, tapping into data characteristics for enhanced model performance. Furthermore, we also embrace hybrid data augmentation techniques, going beyond conventional approaches by further generating customized noise maps to enhance recognition of complex components. Through extensive experiments, our proposed denoising model consistently outperforms meticulously calibrated traditional filters on average. Additionally, when juxtaposing our physics-informed loss with the conventional DnCNN loss across different dataset sizes, our approach consistently emerges superior, particularly with more constrained datasets. Furthermore, the new loss shows effectiveness in general image denoising, which proves the robustness and universality of this suggested method. These findings underscore the promise of our method as a general avenue to seamlessly weave domain expertise into neural networks, especially in data-scarce situations of most industr

关键词： Gaussian noise (electronic)

来源：评论

学校读者我要写书评

暂无评论

Metaheuristic Optimization Through Deep Learning Classification of COVID-19 in Chest X-Ray Images

引用

Computers, Materials & Continua 2022年第11期73卷 4193-4210页

作者： Nagwan Abdel Samee El-Sayed M.El-Kenawy Ghada Atteia Mona M.Jamjoom Abdelhameed Ibrahim Abdelaziz A.Abdelhamid Noha E.El-Attar Tarek Gaber Adam Slowik Mahmoud Y.Shams Department of Information Technology College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh 11671Saudi Arabia Department of Communications and Electronics Delta Higher Institute of Engineering and TechnologyMansoura 35111Egypt Faculty of Artificial Intelligence Delta University for Science and TechnologyMansoura 35712Egypt Department of Computer Sciences College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh 11671Saudi Arabia Computer Engineering and Control Systems Department Faculty of EngineeringMansoura University35516MansouraEgypt Department of Computer Science Faculty of Computer and Information SciencesAin Shams University11566CairoEgypt Department of Computer Science College of Computing and Information TechnologyShaqra University11961Saudi Arabia Faculty of Computers and Artificial Intelligence Benha UniversityEgypt School of Science Engineeringand EnvironmentUniversity of SalfordUK Faculty of Computers and Informatics Suez Canal UniversityIsmailia41522Egypt Koszalin University of Technology Poland Faculty of Artificial Intelligence Kafrelsheikh UniversityKafrelsheikh33511Egypt

As corona virus disease(COVID-19)is still an ongoing global outbreak,countries around the world continue to take precautions and measures to control the spread of the *** of the excessive number of infected patients and the resulting deficiency of testing kits in hospitals,a rapid,reliable,and automatic detection of COVID-19 is in extreme need to curb the number of *** analyzing the COVID-19 chest X-ray images,a novel metaheuristic approach is proposed based on hybrid dipper throated and particle swarm *** lung region was segmented from the original chest X-ray images and augmented using various transformation ***,the augmented images were fed into the VGG19 deep network for feature *** the other hand,a feature selection method is proposed to select the most significant features that can boost the classification ***,the selected features were input into an optimized neural network for *** neural network is optimized using the proposed hybrid *** experimental results showed that the proposed method achieved 99.88%accuracy,outperforming the existing COVID-19 detection *** addition,a deep statistical analysis is performed to study the performance and stability of the proposed *** results confirm the effectiveness and superiority of the proposed approach.

关键词： Covid-19 feature selection dipper throated optimization particle swarm optimization deep learning

来源：评论

学校读者我要写书评

暂无评论

Interaction-Aware Cut-In Trajectory Prediction and Risk Assessment in Mixed Traffic

引用

IEEE/CAA Journal of Automatica Sinica 2022年第10期9卷 1752-1762页

作者： Xianglei Zhu Wen Hu Zejian Deng Jinwei Zhang Fengqing Hu Rui Zhou Keqiu Li Fei-Yue Wang the College of Intelligence and Computing Tianjin UniversityTianjin 300350 the China Automotive Technology and Research Center Co.Ltd. Tianjin 300300China the State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body College of Mechanical and Vehicle EngineeringHunan UniversityChangsha 410082China the Department of Mechanical and Mechatronics Engineering University of WaterlooWaterlooON N2L3G1Canada the School of Mechanical Engineering Beijing Institute of TechnologyBeijing 100081China the Macao University of Science and Technology MacaoChina the Waytous Inc. Qingdao 266000China IEEE the State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China

Accurately predicting the trajectories of surrounding vehicles and assessing the collision risks are essential to avoid side and rear-end collisions caused by *** improve the safety of autonomous vehicles in the mixed traffic,this study proposes a cut-in prediction and risk assessment method with considering the interactions of multiple traffic *** integration of the support vector machine and Gaussian mixture model(SVM-GMM)is developed to simultaneously predict cut-in behavior and *** dimension of the input features is reduced through Chebyshev fitting to improve the training efficiency as well as the online inference *** on the predicted trajectory of the cut-in vehicle and the responsive actions of the autonomous vehicles,two risk measurements are introduced to formulate the comprehensive interaction risk through the combination of Sigmoid function and Softmax ***,the comparative analysis is performed to validate the proposed method using the naturalistic driving *** results show that the proposed method can predict the trajectory with higher precision and effectively evaluate the risk level of a cut-in maneuver compared to the methods without considering interaction.

关键词： Cut-in behavior interaction-aware mixed traffic risk assessment trajectory prediction

来源：评论

学校读者我要写书评

暂无评论

Multistability Analysis and Adaptive Feedback control on a New Financial Risk System

引用

International Journal of Applied and Computational Mathematics 2023年第5期9卷 1-16页

作者： Johansyah, Muhamad Deni Sambas, Aceng Vaidyanathan, Sundarapandian Benkouider, Khaled Sulaiman, Ibrahim Mohammed Department of Mathematics Universitas Padjadjaran Jatinangor Sumedang 45363 Indonesia Faculty of Informatics and Computing Universiti Sultan Zainal Abidin Campus Besut Terengganu Besut 22200 Malaysia Department of Mechanical Engineering Universitas Muhammadiyah Tasikmalaya Tamansari Gobras Tasikmalaya 46196 Indonesia Center for Control System Vel Tech University Avaddi Chennai 60062 India Department of Non-Destructive Automatic Testing Laboratory Jijel University Rue Rouikha Mokhta Jijel 1800 Algeria School of Quantitative Science University Utara Malaysia Sintok Kedah Bukit Kayu Hitam 06010 Malaysia

This study presents the findings of a new financial risk system with one absolute function and two quadratic nonlinearities. The rich dynamic behaviors of the new financial risk system are discussed analytically and numerically through stability analysis, phase portrait, bifurcation diagram, Lyapunov exponent, multistability and coexisting attractors. In addition, a new financial risk system with two quadratic nonlinearities and one absolute function shows multistability and coexistence of chaotic attractors. Using adaptive feedback control, we established a new control law for fully synchronising the proposed chaotic financial systems with itself. To conclude, some numerical simulation was presented using MATLAB to illustrate the efficiency of the designed method. © 2023, The Author(s), under exclusive licence to Springer Nature India Private Limited.

关键词： Backstepping control Chaos Financial chaotic system Synchronisation

来源：评论

学校读者我要写书评

暂无评论

Estimating infinitesimal generators of stochastic systems with formal error bounds: A data-driven approach 24

Estimating infinitesimal generators of stochastic systems wi...

引用

24th ACM International Conference on Hybrid Systems Computation and control, HSCC 2021, held as part of the 14th Cyber Physical Systems and Internet-of-Things Week, CPS-IoT Week 2021

作者： Lavaei, Abolfazl Nejati, Ameneh Soudjani, Sadegh Zamani, Majid Institute for Dynamic Systems and Control Eth Zurich Switzerland Department of Electrical Engineering Technical University of Munich Germany School of Computing Newcastle University United Kingdom Department of Computer Science University of Colorado Boulder United States

ISBN: (纸本)9781450383394

In this work, we propose a data-driven technique for a formal estimation of infinitesimal generators of continuous-time stochastic systems with unknown dynamics. In the proposed framework, we first approximate the infinitesimal generator of the solution process via a set of data collected from solution processes of unknown systems. We then put some proper assumptions on dynamics of systems and quantify the closeness between the infinitesimal generator and its approximation while providing a priori guaranteed confidence bound. We show that both the time discretization and the number of data play significant roles in providing a reasonable closeness precision. Motivations and Related Works. Infinitesimal generator of a stochastic process (i.e., a partial differential operator that encodes a great deal of information about the process) plays a principal role in many analysis frameworks including: (i) stability analysis of continuous-time stochastic systems via Lyapunov functions [1], (ii) establishing similarity relations between two continuous-time stochastic systems via stochastic simulation functions [2], (iii) constructing finite abstractions of continuous-time stochastic control systems [3], and (iv) safety analysis of continuous-time stochastic systems via barrier certificates [4], to name a few. In the relevant literature, only [5] studies the estimation of infinitesimal generators, whose approach is based on the assumption of knowing the precise model and discretizing both time and state. To the best of our knowledge, our work is the first to propose a method for estimating the infinitesimal generator of stochastic systems with unknown dynamics while providing formal guarantees. The complete version of this work can be found in [6]. © 2021 Owner/Author.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Video2Reward: Generating Reward Function from Videos for Legged Robot Behavior Learning

arXiv

引用

arXiv 2024年

作者： Zeng, Runhao Zhou, Dingjie Liang, Qiwei Liu, Junlin Li, Hui Huang, Changxin Li, Jianqiang Hu, Xiping Sun, Fuchun Artificial Intelligence Research Institute Shenzhen MSU-BIT University China College of Mechatronics and Control Engineering Shenzhen University China College of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Department of Computer Science and Technology Tsinghua University China

Learning behavior in legged robots presents a significant challenge due to its inherent instability and complex constraints. Recent research has proposed the use of a large language model (LLM) to generate reward functions in reinforcement learning, thereby replacing the need for manually designed rewards by experts. However, this approach, which relies on textual descriptions to define learning objectives, fails to achieve controllable and precise behavior learning with clear directionality. In this paper, we introduce a new video2reward method, which directly generates reward functions from videos depicting the behaviors to be mimicked and learned. Specifically, we first process videos containing the target behaviors, converting the motion information of individuals in the videos into keypoint trajectories represented as coordinates through a video2text transforming module. These trajectories are then fed into an LLM to generate the reward function, which in turn is used to train the policy. To enhance the quality of the reward function, we develop a video-assisted iterative reward refinement scheme that visually assesses the learned behaviors and provides textual feedback to the LLM. This feedback guides the LLM to continually refine the reward function, ultimately facilitating more efficient behavior learning. Experimental results on tasks involving bipedal and quadrupedal robot motion control demonstrate that our method surpasses the performance of state-of-the-art LLM-based reward generation methods by over 37.6% in terms of human normalized score. More importantly, by switching video inputs, we find our method can rapidly learn diverse motion behaviors such as walking and running. Copyright © 2024, The Authors. All rights reserved.

关键词： Robots

来源：评论

学校读者我要写书评

暂无评论

Predicting Central Lymph Node Metastasis in Papillary Thyroid Carcinoma Using Multimodal Data 8th

Predicting Central Lymph Node Metastasis in Papillary Thyroi...

引用

8th International Conference on Life System Modeling and Simulation, LSMS 2024 and 8th International Conference on Intelligent computing for Sustainable Energy and Environment, ICSEE 2024

作者： Shao, Rui Ben, JiangYuan Lin, Pengcheng Ge, Shu Wu, ChengGang Zhang, Kun He, Ying School of Electrical Engineering and Automation Nantong University Nantong China Department of Ultrasound Tumor Hospital Affiliated to Nantong University Nantong China Nantong Key Laboratory of Intelligent Control and Intelligent Computing Nantong China Nantong Institute of Technology Jiangsu Nantong China

ISBN: (纸本)9789819601875

During thyroid cancer surgery, the debate continues regarding whether to perform routine prophylactic central lymph node dissection on the same side as the lesion. Central lymph node metastasis has been proven to be clinically significant in guiding treatment decisions. Therefore, accurately predicting central lymph node metastasis preoperatively is an significant step in the diagnosis and treatment of thyroid cancer. In clinical practice, ultrasound examination is commonly used to assess thyroid cancer. However, due to the low contrast and high noise of ultrasound images, the accuracy of ultrasound in identifying lymph node metastasis is limited. This study proposes a preoperative prediction of lymph node metastasis based on MFF-Resnet, which is based on Resnet50. This paper constructs multimodal data by combining deep learning with radiomics and clinical factors, and uses auxiliary branches to incorporate original information and low-level features of ultrasound images for prediction tasks. The AUC of MFF-Resnet reaches 0.8576, with an accuracy of 0.8018, and an F1 score of 0.8281. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Diseases

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：