检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Alawad, Mohammed Gao, Shang Shekar, Mayanka Chandra Hasan, S.M. Shamimul Christian, J. Blair Wu, Xiao-Cheng Durbin, Eric B. Doherty, Jennifer Stroup, Antoinette Coyle, Linda Penberthy, Lynne Tourassi, Georgia Electrical and Computer Engineering Department Wayne State University DetroitMI United States Computational Sciences and Engineering Division Oak Ridge National Laboratory Oak RidgeTN United States Louisiana Tumor Registry Louisiana State University Health Sciences Center New OrleansLA United States Kentucky Cancer Registry and College of Medicine University of Kentucky LexingtonKY United States Utah Cancer Registry University of Utah School of Medicine Salt Lake CityUT United States New Jersey State Cancer Registry New Jersey Department of Health TrentonNJ United States Information Management Services Inc CalvertonMD United States Surveillance Research Program Division of Cancer Control and Population Sciences National Cancer Institute BethesdaMD United States National Center for Computational Sciences Oak Ridge National Laboratory Oak RidgeTN United States

A key component of deep learning (DL) for natural language processing (NLP) is word embeddings. Word embeddings that effectively capture the meaning and context of the word that they represent can significantly improve the performance of downstream DL models for various NLP tasks. Many existing word embeddings techniques capture the context of words based on word co-occurrence in documents and text;however, they often cannot capture broader domain-specific relationships between concepts that may be crucial for the NLP task at hand. In this paper, we propose a method to integrate external knowledge from medical terminology ontologies into the context captured by word embeddings. Specifically, we use a medical knowledge graph, such as the unified medical language system (UMLS), to find connections between clinical terms in cancer pathology reports. This approach aims to minimize the distance between connected clinical concepts. We evaluate the proposed approach using a Multitask Convolutional Neural Network (MT-CNN) to extract six cancer characteristics – site, subsite, laterality, behavior, histology, and grade – from a dataset of 900K cancer pathology reports. The results show that the MT-CNN model which uses our domain informed embeddings outperforms the same MT-CNN using standard word2vec embeddings across all tasks, with an improvement in the overall micro- and macro-F1 scores by 4.97%and 22.5%, respectively. Copyright © 2021, The Authors. All rights reserved.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Hierarchic neighbors embedding

arXiv

引用

arXiv 2019年

作者： Liu, Shenglan Yu, Yang Liu, Yang Qiao, Hong Feng, Lin Feng, Jiashi School of Computer Science and Technology Faculty of Electronic Information and Electrical Engineering Dalian University of Technology Dalian Liaoning116024 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Department of Electrical and Computer Engineering National University of Singapore Singapore Singapore

Manifold learning now plays a very important role in machine learning and many relevant applications. Although its superior performance in dealing with nonlinear data distribution, data sparsity is always a thorny knot. There are few researches to well handle it in manifold learning. In this paper, we propose Hierarchic Neighbors Embedding (HNE), which enhance local connection by the hierarchic combination of neighbors. After further analyzing topological connection and reconstruction performance, three different versions of HNE are given. The experimental results show that our methods work well on both synthetic data and high-dimensional real-world tasks. HNE develops the outstanding advantages in dealing with general data. Furthermore, comparing with other popular manifold learning methods, the performance on sparse samples and weak-connected manifolds is better for HNE. Copyright © 2019, The Authors. All rights reserved.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

A lyapunov-based small-gain theorem for infinite networks

arXiv

引用

arXiv 2019年

作者： Kawan, Christoph Mironchenko, Andrii Swikir, Abdalla Noroozi, Navid Zamani, Majid Institute of Informatics Ludwig Maximilian University of Munich Germany Faculty of Computer Science and Mathematics University of Passau Passau94032 Germany Department of Electrical and Computer Engineering Technical University of Munich Germany Otto-von-Guericke University Magdeburg Laboratory for Systems Theory and Automatic Control Magdeburg Germany Computer Science Department University of Colorado BoulderCO80309 United States Computer Science Department Ludwig Maximilian University of Munich Germany

This paper presents a small-gain theorem for networks composed of a countably infinite number of finite-dimensional subsystems. Assuming that each subsystem is exponentially input-to-state stable, we show that if the gain operator, collecting all the information about the internal Lyapunov gains, has a spectral radius less than one, the overall infinite network is exponentially input-to-state stable. The effectiveness of our result is illustrated through several examples including nonlinear spatially invariant systems with sector nonlinearities and a road traffic *** Codes 37B25, 37L15, 93D05, 93A15 Copyright © 2019, The Authors. All rights reserved.

关键词： Large scale systems

来源：评论

学校读者我要写书评

暂无评论

A General Framework for Flexible Multi-Cue Photometric Point Cloud Registration

A General Framework for Flexible Multi-Cue Photometric Point...

引用

IEEE International Conference on Robotics and Automation

作者： Bartolomeo Della Corte Igor Bogoslavskyi Cyrill Stachniss Giorgio Grisetti Department of Computer Control and Management Engineering Antonio Ruberti Rome Italy

The ability to build maps is a key functionality for the majority of mobile robots. A central ingredient to most mapping systems is the registration or alignment of the recorded sensor data. In this paper, we present a general methodology for photometric registration that can deal with multiple different cues. We provide examples for registering RGBD as well as 3D LIDAR data. In contrast to popular point cloud registration approaches such as ICP our method does not rely on explicit data association and exploits multiple modalities such as raw range and image data streams. Color, depth, and normal information are handled in an uniform manner and the registration is obtained by minimizing the pixel-wise difference between two multi-channel images. We developed a flexible and general framework and implemented our approach inside that framework. We also released our implementation as open source C++ code. The experiments show that our approach allows for an accurate registration of the sensor data without requiring an explicit data association or model-specific adaptations to datasets or sensors. Our approach exploits the different cues in a natural and consistent way and the registration can be done at framerate for a typical range or imaging sensor.

关键词： Three-dimensional displays Robot sensing systems Cameras Iterative closest point algorithm Minimization Integrated circuit modeling Laser radar

来源：评论

学校读者我要写书评

暂无评论

Study on Insulator Recognition Method Based on Simulated Samples Expansion

Study on Insulator Recognition Method Based on Simulated Sam...

引用

第30届中国控制与决策会议

作者： Haiyan Cheng Rui Chen Jinna Wang Xinyue Liu Muliu Zhang Yongjie Zhai North China Electric Power University Department of Control and Computer Engineering

In the application of the deep learning to the unmanned aerial vehicle(UAV) autonomous inspection, a problem about the insufficiency of both quantity and quality for insulator images emerges. In the light of this situation, a sample expansion method based on a combination of statute and 3 D modelling technology is proposed. Its feasibility is verified in a deep convolutional neural network by using five kinds of insulator simulated samples. The classification accuracy acquired by the proposed method is higher verified by the comparison of the experimental results, which proves its superiority to the traditional method containing no simulated samples. It is concluded that the simulated intensive samples of pure background have a significant effect on the accuracy of network classification. And when the ratio of real samples to simulated intensive samples goes to an appropriate value, the classification has the best accuracy.

关键词： 3D modelling Simulated intensive samples Deep learning Insulators

来源：评论

学校读者我要写书评

暂无评论

Deep differentiable random forests for age estimation

arXiv

引用

arXiv 2019年

作者： Shen, Wei Guo, Yilu Wang, Yan Zhao, Kai Wang, Bo Yuille, Alan Department of Computer Science Johns Hopkins University BaltimoreMD21218-2608 United States School of Communication and Information Engineering Shanghai University Shanghai200444 China College of Computer and Control Engineering Nankai University Tianjin300071 China Vector Institute and Peter Munk Cardiac Center of University Health Network TorontoON Canada

Age estimation from facial images is typically cast as a label distribution learning or regression problem, since aging is a gradual progress. Its main challenge is the facial feature space w.r.t. ages is inhomogeneous, due to the large variation in facial appearance across different persons of the same age and the non-stationary property of aging. In this paper, we propose two Deep Differentiable Random Forests methods, Deep Label Distribution Learning Forest (DLDLF) and Deep Regression Forest (DRF), for age estimation. Both of them connect split nodes to the top layer of convolutional neural networks (CNNs) and deal with inhomogeneous data by jointly learning input-dependent data partitions at the split nodes and age distributions at the leaf nodes. This joint learning follows an alternating strategy: (1) Fixing the leaf nodes and optimizing the split nodes and the CNN parameters by Back-propagation;(2) Fixing the split nodes and optimizing the leaf nodes by Variational Bounding. Two Deterministic Annealing processes are introduced into the learning of the split and leaf nodes, respectively, to avoid poor local optima and obtain better estimates of tree parameters free of initial values. Experimental results show that DLDLF and DRF achieve state-of-the-art performance on three age estimation datasets. Copyright © 2019, The Authors. All rights reserved.

关键词： Regression analysis

来源：评论

学校读者我要写书评

暂无评论

Particle swarm optimized command filtered backstepping control for an active magnetic bearing system

Particle swarm optimized command filtered backstepping contr...

引用

2018 IEEE International Conference on Information and Automation, ICIA 2018

作者： Chen, Ziyin Lin, Zhe Yue, Chunyu Li, Yang Department of Control Engineering Beijing Instituste of Space Mechanics Electric Beijing Haidian100094 China Faculty of Computer Science Otto-von-Guericke-University Magdeburg Magdeburg39106 Germany

ISBN: (纸本)9781538680698

This paper discusses the position control problem for the active magnetic bearing (AMB) suspension system under external disturbance. The command filtered approach is applied for the controller design to reduce the complexity for repeatedly calculating the analytic derivatives of the virtual control with less computation effort compared to standard backstepping. The adaptive law and robust terms are obtained through Lyapunov theory and asymptotic stability can be guaranteed for the filtered tracking errors. The particle swarm optimization (PSO) algorithm is adopted to adjust the control parameters with guaranteed dynamic performance. Finally, the simulation results illustrate the tracking performance can be made arbitrarily close to the solution of standard backstepping. © 2018 IEEE.

关键词： Backstepping

来源：评论

学校读者我要写书评

暂无评论

Blockchain Technology Based Decentralized Energy Trading for Multiple-Microgrid Systems

Blockchain Technology Based Decentralized Energy Trading for...

引用

IEEE Conference on Energy Internet and Energy System Integration (EI2)

作者： Ting Li Wei Zhang Ning Chen Minhui Qian Ying Xu School of Electrical Engineering and Automation Harbin Institute of Technology Harbin China State Key Laboratory of Operation and Control of Renewable Energy & Storage Systems China EPRI Nanjing China Department of Electrical Engineering & Computer Science University of Central Florida Orlando USA

ISBN: (数字)9781728131375

ISBN: (纸本)9781728131382

With the emerging technology for distributed generation and urge of improving quality of service of power supply for energy users, more and more Microgrids (MGs) are integrated into the distributed networks to serve the energy users. These Microgrids are gradually formulating a Multi-Microgrid System (MMGS, Multi-Microgrid System), which will play an important role for the future energy supply. Building a centralized control center not only increases the expense of investment, but also brings the issues of maintaining fairness among energy users. To address these problems, this paper proposes a peer-to-peer method for energy trading of MMGS, based on the idea of decentralized trading. An auction-based trading mechanism suitable for peer-to-peer energy trading is proposed first. For this mechanism, the MGs firstly declare their energy buying bids or selling quotations. Then, the market-clearing price is determined by using a unified weight clearing algorithm in a decentralized manner. By applying the edge technology of Blockchain, the implementation of the proposed peer-to-peer energy trading method, including the architecture, procedures, security check, etc., is also discussed. The proposed Blockchain-based energy trading platform can realize the decentralized and autonomous energy trading of MGs within an MMGS. A case study with an MMGS with 10 MG units is provided to demonstrate the effectiveness of the proposed approach.

关键词： Peer-to-peer computing Microgrids Blockchain Distributed power generation Electronic commerce Power distribution economics Pricing

来源：评论

学校读者我要写书评

暂无评论

Performance–guaranteed consensus control inspired by the mammalian limbic system for a class of nonlinear multi-agents ⁎

引用

IFAC-PapersOnLine 2020年第2期53卷 9496-9501页

作者： Ignacio Rubio Scola Luis Rodolfo Garcia Carrillo João P. Hespanha Rogelio Lozano CIFASIS (CONICET-UNR) and Department of Mathematics National University of Rosario Rosario 2000 Argentina. Unmanned Systems Laboratory. School of Engineering and Computer Sciences Texas A&M University - Corpus Christi TexasUSA. 78412-5797. Center for Control Dynamical Systems and ComputationUniversity of California Santa Barbara CA 93106 USA. Heuristique et Diagnostic des Systemes Complexes Universit é de Technologie de Compiègne UMR CNRS 7253 60200 CompiegneFrance

Computational models of emotional learning observed in the mammalian brain have inspired diverse self-learning control approaches. These architectures are promising in terms of their fast learning ability and low computational cost. In this paper, the objective is to establish performance–guaranteed emotional learning–inspired control (ELIC) strategies for autonomous multi–agent systems (MAS), where each agent incorporates an ELIC structure to support the consensus controller. The objective of each ELIC structure is to identify and compensate model differences between the theoretical assumptions taken into account when tuning the consensus protocol, and the real conditions encountered in the real system to be stabilized. Stability of the closed-loop MAS is demonstrated using a Lyapunov analysis. Simulation results based on the consensus task of a group of inverted pendulums demonstrate the effectiveness of the proposed ELIC for stabilization of nonlinear MAS.

关键词： Multi-agents systems Biologically-inspired control Robust control Distributed control Nonlinear control

来源：评论

学校读者我要写书评

暂无评论

Improved DFIG-WPGS Efficiency via the tuned PI-Resonant Unit controller based on BAT Algorithm

Improved DFIG-WPGS Efficiency via the tuned PI-Resonant Unit...

引用

International Conference on computer, control, Electrical, and Electronics engineering (ICCCEEE)

作者： Omar Busati Alzain Xiangiie Liu Xiaobing Kong Alhassan Busati Alzain Mohammed.A. Abdalla Dhaifallah Ahmed Ali School of Control and Computer Eng. North China Electric Power Univ. Beijing China Department of Control Engineering Karary Univ. Khartoum Sudan School of E.Power Eng. North China Electric Power Univ. Beijing China

ISBN: (数字)9781728110066

ISBN: (纸本)9781728110073

This study introduces rotor-edge current control concerning stator-edge power behavior based on the online tuned Resonant Proportional-Integral Regulator using Discrete-BAT optimization Algorithm (DPIR-ON-DBAT) for the DFIG-WPGS as a Recurrent Artificial Neural Network (RANN). The DPIR-ON-DBAT is designed to control the distorted system with several dynamic positions and control goals. The +dq rotor-edge current components, swings of electromagnetic torque and stator-edge power, are regulated by the proposed control with a resonant part oscillated at 6-times the main system frequency. Consequently, these variables are instantly regulated by the DPIR-ON-DBAT technique without extra succession disassembling the segments. The simulation results are implemented using Matlab-Simulink for the introduced control of a 2MW DFIG-WPGS, which gave a better dynamic behavior than the customary PIR-Naslin and PI control principle.

关键词： Doubly fed induction generators Voltage control Mathematical model Power systems Heuristic algorithms Stator windings

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：