检索结果-内蒙古大学图书馆

arXiv 2019年

作者： Karttunen, Janne Kanervisto, Anssi Hautamäki, Ville Kyrki, Ville School of Computing University of Eastern Finland Joensuu Finland Department of Electrical Engineering and Automation Aalto University Espoo Finland

Training agents with reinforcement learning based techniques requires thousands of steps, which translates to long training periods when applied to robots. By training the policy in a simulated environment we avoid such limitation. Typically, the action spaces in a simulation and real robot are kept as similar as possible, but if we want to use a generic simulation environment, this strategy will not work. Video games, such as Doom (1993), offer a crude but multi-purpose environments that can used for learning various tasks. However, original Doom has four discrete actions for movement and the robot in our case has two continuous actions. In this work, we study the transfer between these two different action spaces. We begin with experiments in a simulated environment, after which we validate the results with experiments on a real robot. Results show that fine-tuning initially learned network parameters leads to unreliable results, but by keeping most of the neural network frozen we obtain above 90% success rate in simulation and real robot experiments. Copyright © 2019, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

On Design of H∞ Structured Controller for Decentralized Control systems

On Design of H∞ Structured Controller for Decentralized Con...

引用

Control Conference (ANZCC), Australian and New Zealand

作者： Yanpeng Guan Wei Xing Zheng Department of Automation Shanxi University Taiyuan China School of Computing Engineering and Mathematics Western Sydney University Sydney Australia

In this paper the problem of H ∞ structured controller design is addressed. The considered problem arises when selection of actuators in decentralized control systems leads to that some rows of the controller gain matrix should be zero. However, the design of controllers with sparse structure is essentially a combinatorial problem that is hard to handle. Towards this end, this paper adopts a mixed-integer programming method to tackle the resulting row-cardinality constraint. The direct mixed-integer programming method is utilized to assist in designing row-sparse H ∞ state feedback controllers for large-scale distributed systems.

关键词： Sparse matrices State feedback Programming Decentralized control Actuators Australia Optimization

来源：评论

学校读者我要写书评

暂无评论

Channel pruning via automatic structure search

arXiv

引用

arXiv 2020年

作者： Lin, Mingbao Ji, Rongrong Zhang, Yuxin Zhang, Baochang Wu, Yongjian Tian, Yonghong Media Analytics and Computing Laboratory Department of Artificial Intelligence School of Informatics Xiamen University China School of Automation Science and Electrical Engineering Beihang University China Co. Ltd China School of Electronics Engineering and Computer Science Peking University Beijing China Peng Cheng Laboratory Shenzhen China

Channel pruning is among the predominant approaches to compress deep neural networks. To this end, most existing pruning methods focus on selecting channels (filters) by importance/ optimization or regularization based on ruleof- thumb designs, which defects in sub-optimal pruning. In this paper, we propose a new channel pruning method based on artificial bee colony algorithm (ABC), dubbed as ABCPruner, which aims to efficiently find optimal pruned structure, i.e., channel number in each layer, rather than selecting "important" channels as previous works did. To solve the intractably huge combinations of pruned structure for deep networks, we first propose to shrink the combinations where the preserved channels are limited to a specific space, thus the combinations of pruned structure can be significantly reduced. And then, we formulate the search of optimal pruned structure as an optimization problem and integrate the ABC algorithm to solve it in an automatic manner to lessen human interference. ABCPruner has been demonstrated to be more effective, which also enables the fine-tuning to be conducted efficiently in an end-to-end manner. Experiments on CIFAR-10 show that ABCPruner reduces 73.68% of FLOPs and 88.68% of parameters with even 0.06% accuracy improvement for VGGNet-16. On ILSVRC-2012, it achieves a reduction of 62.87% FLOPs and removes 60.01% of parameters with negligible accuracy cost for ResNet-152. The source codes can be available at https://***/lmbxmu/ABCPruner. Copyright © 2020, 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

1D convolutional neural network models for sleep arousal detection

arXiv

引用

arXiv 2019年

作者： Zabihi, Morteza Rad, Ali Bahrami Kiranyaz, Serkan Särkkä, Simo Gabbouj, Moncef Department of Computing Sciences Tampere University Tampere Finland Department of Electrical Engineering and Automation Aalto University Espoo Finland Electrical Engineering Department Qatar University Doha Qatar

Sleep arousals transition the depth of sleep to a more superficial stage. The occurrence of such events is often considered as a protective mechanism to alert the body of harmful stimuli. Thus, accurate sleep arousal detection can lead to an enhanced understanding of the underlying causes and influencing the assessment of sleep quality. Previous studies and guidelines have suggested that sleep arousals are linked mainly to abrupt frequency shifts in EEG signals, but the proposed rules are shown to be insufficient for a comprehensive characterization of arousals. This study investigates the application of five recent convolutional neural networks (CNNs) for sleep arousal detection and performs comparative evaluations to determine the best model for this task. The investigated state-of-the-art CNN models have originally been designed for image or speech processing. A detailed set of evaluations is performed on the benchmark dataset provided by PhysioNet/computing in Cardiology Challenge 2018, and the results show that the best 1D CNN model has achieved an average of 0.31 and 0.84 for the area under the precision-recall and area under the ROC curves, respectively. Copyright © 2019, The Authors. All rights reserved.

关键词： Sleep research

来源：评论

学校读者我要写书评

暂无评论

Potential of multimodal large language models for data mining of medical images and free-text reports

Meta-Radiology

引用

Meta-Radiology 2024年第4期2卷

作者： Zhang, Yutong Pan, Yi Zhong, Tianyang Dong, Peixin Xie, Kangni Liu, Yuxiao Jiang, Hanqi Wu, Zihao Liu, Zhengliang Zhao, Wei Zhang, Wei Zhao, Shijie Zhang, Tuo Jiang, Xi Shen, Dinggang Liu, Tianming Zhang, Xin Institute of Medical Research Northwestern Polytechnical University Xi'an 710072 China School of Computing The University of Georgia Athens 30602 United States School of Life Science and Technology University of Electronic Science and Technology of China Chengdu 611731 China School of Automation Northwestern Polytechnical University Xi'an 710072 China School of Electronics and Information Northwestern Polytechnical University Xi'an 710072 China School of Biomedical Engineering ShanghaiTech University Shanghai 201210 China Department of Radiology The Second Xiangya Hospital Central South University Changsha 410011 China Clinical Research Center for Medical Imaging in Hunan Province Changsha 410011 China Institute of Biomedical and Health Engineering Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen 518055 China School of Computer and Cyber Sciences Augusta University Augusta 30912 United States Shanghai United Imaging Intelligence Co. Ltd. Shanghai 200230 China Shanghai Clinical Research and Trial Center Shanghai 201210 China

Medical images and radiology reports are essential for physicians to diagnose medical conditions. However, the vast diversity and cross-source heterogeneity inherent in these data have posed significant challenges to the generalizability of current data-mining methods for clinical decision-making. Recently, multimodal large language models (MLLMs), especially Gemini-Vision-series (Gemini) and GPT-4-series (GPT-4) models, have revolutionized numerous domains, significantly impacting the medical field. In this study, we conducted a detailed evaluation of the performance of the Gemini series models (including Gemini-1.0-Pro-Vision, Gemini-1.5-Pro, and Gemini-1.5-Flash) and GPT series models (including GPT-4o, GPT-4-Turbo, and GPT-3.5-Turbo) across 14 medical datasets, covering 5 medical imaging categories (dermatology, radiology, dentistry, ophthalmology, and endoscopy) and 3 radiology report datasets. The investigated tasks encompass disease classification, lesion segmentation, anatomical localization, disease diagnosis, report generation, and lesion detection. Moreover, we also validated the performance of the Claude-3-Opus, Yi-Large, Yi-Large-Turbo, and LLaMA 3 models to gain a comprehensive understanding of the MLLM models in the medical field. Our experimental results demonstrated that Gemini-series models excelled in report generation and lesion detection but faces challenges in disease classification and anatomical localization. Conversely, GPT-series models exhibited proficiency in lesion segmentation and anatomical localization but encountered difficulties in disease diagnosis and lesion detection. Additionally, both the Gemini series and GPT series contain models that have demonstrated commendable generation efficiency. While both models hold promise in reducing physician workload, alleviating pressure on limited healthcare resources, and fostering collaboration between clinical practitioners and artificial intelligence technologies, substantial enhancements

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-modal remote sensory learning for multi-objects over autonomous devices

引用

Frontiers in bioengineering and biotechnology 2025年 13卷 1430222页

作者： Aysha Naseer Naif Almudawi Hanan Aljuaid Abdulwahab Alazeb Yahay AlQahtani Asaad Algarni Ahmad Jalal Hui Liu Department of Computer Science Air University Islamabad Pakistan. Department of Information Systems College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University Riyadh Saudi Arabia. Department of Computer Science College of Computer Science and Information System Najran University Najran Saudi Arabia. Department of Informatics and Computer Systems King Khalid University Abha Saudi Arabia. Department of Computer Sciences Faculty of Computing and Information Technology Northern Border University Rafha Saudi Arabia. Department of Computer Science and Engineering College of Informatics Korea University Seoul South Korea. Guodian Nanjing Automation Co. Ltd. Nanjing China. Jiangsu Key Laboratory of Intelligent Medical Image Computing School of Future Technology Nanjing University of Information Science and technology Nanjing China. Cognitive Systems Lab University of Bremen Bremen Germany.

Introduction:There has been an increasing focus on object segmentation within remote sensing images in recent years due to advancements in remote sensing technology and the growing significance of these images in both military and civilian realms. In these situations, it is critical to accurately and quickly identify a wide variety of objects. In many computer vision applications, scene recognition in aerial-based remote sensing imagery presents a common issue. Method:However, several challenging elements make this work especially difficult: (i) Different objects have different pixel densities; (ii) objects are not evenly distributed in remote sensing images; (iii) objects can appear differently depending on viewing angle and lighting conditions; and (iv) there are fluctuations in the number of objects, even the same type, in remote sensing images. Using a synergistic combination of Markov Random Field (MRF) for accurate labeling and Alex Net model for robust scene recognition, this work presents a novel method for the identification of remote sensing objects. During the labeling step, the use of MRF guarantees precise spatial contextual modeling, which improves comprehension of intricate interactions between nearby aerial objects. By simultaneously using deep learning model, the incorporation of Alex Net in the following classification phase enhances the model's capacity to identify complex patterns in aerial images and adapt to a variety of object attributes. Results:Experiments show that our method performs better than others in terms of classification accuracy and generalization, indicating its efficacy analysis on benchmark datasets such as UC Merced Land Use and AID. Discussion:Several performance measures were calculated to assess the efficacy of the suggested technique, including accuracy, precision, recall, error, and F1-Score. The assessment findings show a remarkable recognition rate of around 97.90% and 98.90%, on the AID and the UC Merced Land datasets,

关键词： autonomous devices deep learning computer vision likelihood estimation multi-modal multi-objects posterior probability remote sensing scene analysis

来源：评论

学校读者我要写书评

暂无评论

Robust multi-objective control for vehicle active suspension systems with structured uncertainties

Robust multi-objective control for vehicle active suspension...

引用

Chinese Control Conference (CCC)

作者： Zhong Cao Sheng-Guo Wang Zhaohui Chen Xiaorong Hou Wenjing Zhao School of Physics and Electronic Engineering Guangzhou University Guangzhou China College of Engineering and College of Computing Informatics University of North Carolina at Charlotte Charlotte NC USA Department of Mathematics and Physics Chongqing University of Science and Technology Chongqing China School of Automation Engineering University of Electronic Science and Technology of China Chengdu China Lab Center Guangzhou University Guangzhou China

ISBN: (数字)9789881563903

ISBN: (纸本)9781728165233

A robust structural control method is proposed to solve the problems of structured uncertainties and multi-objective control for vehicle active suspension systems. The considered model is a quarter model of vehicle subject to unknown ground disturbance and structured uncertainties. The proposed method is based on robust structured control well supported by singular value decomposition (SVD) method, yielding a multi-objective control design approach, applicable to almost any active suspension systems. Some experiments show that the control performance of robust structural control for vehicle active suspension systems with structured uncertainties is significantly improved comparing with passive controlled system and existing other controlled active suspension systems.

关键词： Suspensions Uncertainty Robustness Roads Mathematical model Robust control Control systems

来源：评论

学校读者我要写书评

暂无评论

Focalizing Regions of Biomarker-Relevance Facilitates Biomarker Prediction on Histopathological Images

SSRN

引用

SSRN 2023年

作者： Gan, Jiefeng Wang, Hanchen Yu, Hui He, Zitong Zhang, Wenjuan Ma, Ke Zhu, Lianghui Bai, Yutong Zhou, Zongwei Yullie, Alan Bai, Xiang Wang, Mingwei Yang, Dehua Chen, Yanyan Chen, Guoan Lasenby, Joan Cheng, Chao Wu, Jia Zhang, Jianjun Wang, Xinggang Chen, Yaobing Wang, Guoping Xia, Tian School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China Department of Engineering University of Cambridge Cambridge United Kingdom Computing + Mathematical Sciences Department California Institute of Technology PasadenaCA United States Wuhan Children's Hospital Tongji Medical College China Department of Computer Science Johns Hopkins University United States Department of Pathology Maternal and Child Hospital of Hubei Province Tongji Medical College Huazhong University of Science and Technology Wuhan China School of Electronic Information and Communications Huazhong University of Science and Technology Wuhan China Department of Radiology Tongji Hospital Tongji Medical College Huazhong University of Science and Technology Wuhan China The National Center for Drug Screening Shanghai Institute of Materia Medica Chinese Academy of Sciences Shanghai China Department of Information Management Tongji Hospital Huazhong University of Science and Technology Wuhan China Wuhan Blood Center Wuhan China Baylor College of Medicine HoustonTX United States Department of Radiation Oncology Stanford University School of Medicine Palo AltoCA United States Thoracic/Head and Neck Medical Oncology Translational Molecular Pathology The University of Texas MD Anderson Cancer Center HoustonTX United States Institute of Pathology Tongji Hospital Tongji Medical College Huazhong University of Science and Technology Wuhan China

Image-based AI has thrived as a potential revolutionised tool for predicting the status of molecular biomarkers, which effectively helps group patients with targeted medical treatments. However, methods based on haematoxylin and eosin-stained (H&E) whole-slide images (WSIs), usually accommodate the entire images directly as model inputs, which could be inefficient as many image patches are not informative or even irrelevant to the biomarkers. In this study, we introduced a new concept of region of biomarker-relevance (ROB) to represent the morphological regions most related to a specific biomarker and to serve as a fundamental element for WSI-based biomarker prediction. We embodied the ROB theory with framework titled saliency ROB search (SRS) to effectively predict biomarker status. We exemplified the SRS by analyzing various adenocarcinoma (LUAD) biomarkers, and achieved better performances than the-state-art of AI approaches. This indicates that on the basis of ROB, AI can reach a better molecular biomarker prediction precision from pathological images. © 2023, The Authors. All rights reserved.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Understanding metric-related pitfalls in image analysis validation

arXiv

引用

arXiv 2023年

作者： Reinke, Annika Tizabi, Minu D. Baumgartner, Michael Eisenmann, Matthias Heckmann-Nötzel, Doreen Kavur, A. Emre Rädsch, Tim Sudre, Carole H. Acion, Laura Antonelli, Michela Arbel, Tal Bakas, Spyridon Benis, Arriel Blaschko, Matthew B. Buettner, Florian Cardoso, M. Jorge Cheplygina, Veronika Chen, Jianxu Christodoulou, Evangelia Cimini, Beth A. Collins, Gary S. Farahani, Keyvan Ferrer, Luciana Galdran, Adrian van Ginneken, Bram Glocker, Ben Godau, Patrick Haase, Robert Hashimoto, Daniel A. Hoffman, Michael M. Huisman, Merel Isensee, Fabian Jannin, Pierre Kahn, Charles E. Kainmueller, Dagmar Kainz, Bernhard Karargyris, Alexandros Karthikesalingam, Alan Kenngott, Hannes Kleesiek, Jens Kofler, Florian Kooi, Thijs Kopp-Schneider, Annette Kozubek, Michal Kreshuk, Anna Kurc, Tahsin Landman, Bennett A. Litjens, Geert Madani, Amin Maier-Hein, Klaus Martel, Anne L. Mattson, Peter Meijering, Erik Menze, Bjoern Moons, Karel G.M. Müller, Henning Nichyporuk, Brennan Nickel, Felix Petersen, Jens Rafelski, Susanne M. Rajpoot, Nasir Reyes, Mauricio Riegler, Michael A. Rieke, Nicola Saez-Rodriguez, Julio Sánchez, Clara I. Shetty, Shravya Summers, Ronald M. Taha, Abdel A. Tiulpin, Aleksei Tsaftaris, Sotirios A. van Calster, Ben Varoquaux, Gaël Yaniv, Ziv R. Jäger, Paul F. Maier-Hein, Lena Faculty of Mathematics and Computer Science Heidelberg University Heidelberg Germany Heidelberg Division of Intelligent Medical Systems Germany NCT Heidelberg A Partnership Between DKFZ University Medical Center Heidelberg Germany Heidelberg Division of Medical Image Computing Germany Heidelberg Division of Intelligent Medical Systems Germany MRC Unit for Lifelong Health and Ageing UCL Centre for Medical Image Computing Department of Computer Science University College London London United Kingdom School of Biomedical Engineering and Imaging Science King’s College London London United Kingdom Instituto de Cálculo CONICET – Universidad de Buenos Aires Buenos Aires Argentina Centre for Medical Image Computing University College London London United Kingdom McGill University Montreal Canada Division of Computational Pathology Dept of Pathology & Laboratory Medicine Indiana University School of Medicine IU Health Information and Translational Sciences Building Indianapolis United States University of Pennsylvania Richards Medical Research Laboratories FL7 PhiladelphiaPA United States Department of Digital Medical Technologies Holon Institute of Technology Holon Israel European Federation for Medical Informatics Le Mont-sur-Lausanne Switzerland Center for Processing Speech and Images Department of Electrical Engineering KU Leuven Leuven Belgium partner site Frankfurt/Mainz a partnership between DKFZ and UCT Frankfurt Marburg Germany Heidelberg Germany Goethe University Frankfurt Department of Medicine Germany Goethe University Frankfurt Department of Informatics Germany and Frankfurt Cancer Insititute Germany Department of Computer Science IT University of Copenhagen Copenhagen Denmark Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V. Dortmund Germany Imaging Platform Broad Institute of MIT and Harvard CambridgeMA United States Centre for Statistics in Medicine University of Oxford Oxford United Kingdom Center for Biomedical In

Validation metrics are key for tracking scientific progress and bridging the current chasm between artificial intelligence (AI) research and its translation into practice. However, increasing evidence shows that particularly in image analysis, metrics are often chosen inadequately. While taking into account the individual strengths, weaknesses, and limitations of validation metrics is a critical prerequisite to making educated choices, the relevant knowledge is currently scattered and poorly accessible to individual researchers. Based on a multi-stage Delphi process conducted by a multidisciplinary expert consortium as well as extensive community feedback, the present work provides the first reliable and comprehensive common point of access to information on pitfalls related to validation metrics in image analysis. While focused on biomedical image analysis, the addressed pitfalls generalize across application domains and are categorized according to a newly created, domain-agnostic taxonomy. The work serves to enhance global comprehension of a key topic in image analysis validation. © 2023, CC BY.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Integrate and Fire Neuron Implementation using CMOS Predictive Technology Model for 32nm

Integrate and Fire Neuron Implementation using CMOS Predicti...

引用

Microelectronics Technology and Devices (SBMicro)

作者： Gabriel Maranhão Janaina G. Guimarães Department of Electrical Engineering Universidade de Brasília Brasília Brazil Department of Control Automation and Computing Engineering Universidade Federal de Santa Catarina Blumenau Brazil

The propose of this work is to evolve the studies on neuromorphic circuits by simulating transistors models that look the most with the ones used in commercial process. Since most of data about company models contain a restricted access, some universities provide predictions models to reproduce the real ones. In this paper we present a silicon analog integrate and fire neuron (I&F), proposed by G. Indiveri as a part of a neuromorphic device. Using 32nm CMOS technology simulated in LTspice with BSIM4v4 and applying predictive parameters provides by Predictive Technology Model (PTM), we were able to reduce the source power, to 0.9V, and chip size of the latest design which was implemented using a 180nm CMOS process at 1.8 power supply.

关键词： Neurons Predictive models Logic gates Semiconductor device modeling Transistors Mathematical model Computational modeling

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：