检索结果-内蒙古大学图书馆

IEEE International Conference on High Performance Computing and Communications (HPCC)

作者： Jing Liu Yifu Zhang Yang Liu Pei Yang Cen Chen Ziqi Zhu Zeng Zeng College of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan China Institute for Infocomm Research Singapore

ISBN: (纸本)9781728176499;9781728176505

High reliability is required for many safety-critical and real-time applications. It is difficult to avoid permanent faults during the executing process of tasks. Once such faults are not processed in time, real-time tasks cannot be completed within the specified time, which may lead to catastrophic consequences. In order to deal with the problem of permanent fault tolerance in heterogeneous real-time systems, in this paper, we propose a balanced cost fault-tolerant scheduling algorithm. First, the algorithm uses the task computation time and the average communication time of its successor nodes as the ranking factors when calculating task priority, so that the priority of the task with larger average communication time improved; second, a task cost table is established, and tasks are allocated to appropriate processors according to the values in the task cost table in processor allocation stage. Experimental results show that the proposed algorithm can effectively improve the system reliability and reduce the completion time.

关键词： Fault tolerance Program processors Scheduling algorithms High performance computing Conferences Fault tolerant systems Real-time systems

来源：评论

学校读者我要写书评

暂无评论

引用

作者： Wan-li SONG Information Engineering School Nanjing Xiaozhuang University Key Laboratory of Intelligent Information Processing Nanjing Xiaozhuang University

With the rapid development and all-round popularization of Internet,more and more data is stored in the form of text in the network *** massive data makes the redundancy of text *** is very important to use text similarity technology to remove duplicate ***,how to effectively improve the accuracy and precision of text similarity calculation is an urgent *** this paper,we propose an improved method to calculate sentence semantic *** method uses word2 vec model to get the semantic information of the text,uses k-means algorithm to cluster the above results,then uses word2 vec model for retraining,and finally gets the sentence *** results indicate that the performance of our algorithm is better improved compared with the traditional word2 vec algorithm.

关键词： Sentence similarity Clustering Word2Vec

来源：评论

学校读者我要写书评

暂无评论

Demonstration Guided Actor-Critic Deep Reinforcement Learning for Fast Teaching of Robots in Dynamic Environments

引用

IFAC-PapersOnLine 2020年第5期53卷 271-278页

作者： Liang Gong Te Sun Xudong Li Ke Lin Natalia Díaz-Rodríguez David Filliat Zhengfeng Zhang Junping Zhang School of Mechanical Engineering Shanghai Jiao Tong University Shanghai China Computer Science and System Engineering Department and Inria Flowers Team ENSTA Institut Polytechnique Paris Palaiseau France Shanghai Key Laboratory of Intelligent Information Processing and School of Computer Science Fudan University Shanghai China

Using direct reinforcement learning (RL) to accomplish a task can be very inefficient, especially in robotic configurations where interactions with the environment are lengthy and costly. Instead, learning from expert demonstration (LfD) is an alternative approach to gain better performance in an RL setting, which also greatly improves sample efficiency. We propose a novel demonstration learning framework for actor-critic based algorithms. Firstly, we put forward an environment pre-training paradigm to initialize the model parameters without interacting with the target environment, which effectively avoids the cold start problem in deep RL scenarios. Secondly, we design a general-purpose LfD framework for most of the mainstream actor-critic RL algorithms that include a policy network and a value function like PPO, SAC, TRPO, A3C. Thirdly, we build a dedicated model training platform to perform the human-robot interaction and numerical experimentation. We evaluate the method in six Mujoco simulated locomotion environments and our robot control simulation platform. Results show that several epochs of pre-training can improve the agent’s performance over the early stage of training. Also, the final converged performance of the RL algorithm is also boosted by external demonstration. In general the sample efficiency is improved by 30% with the proposed method. Our demonstration pipeline makes full use of the exploration property of the RL algorithm, which is feasible for fast teaching robots in dynamic environments.

关键词： deep learning deep reinforcement learning learning from demonstration (LfD) actor-critic framework robotics

来源：评论

学校读者我要写书评

暂无评论

Global context-aware progressive aggregation network for salient object detection

arXiv

引用

arXiv 2020年

作者： Chen, Zuyao Xu, Qianqian Cong, Runmin Huang, Qingming University of Chinese Academy of Sciences Beijing China Key Lab. of Intelligent Information Processing ICT CAS Beijing China Institute of Information Science Beijing Jiaotong University Beijing China Key Lab. of Big Data Mining and Knowledge Management CAS Beijing China Peng Cheng Laboratory Shenzhen Guangdong China

Deep convolutional neural networks have achieved competitive performance in salient object detection, in which how to learn effective and comprehensive features plays a critical role. Most of the previous works mainly adopted multiplelevel feature integration yet ignored the gap between different features. Besides, there also exists a dilution process of high-level features as they passed on the top-down pathway. To remedy these issues, we propose a novel network named GCPANet to effectively integrate low-level appearance features, high-level semantic features, and global context features through some progressive context-aware Feature Interweaved Aggregation (FIA) modules and generate the saliency map in a supervised way. Moreover, a Head Attention (HA) module is used to reduce information redundancy and enhance the top layers features by leveraging the spatial and channel-wise attention, and the Self Refinement (SR) module is utilized to further refine and heighten the input features. Furthermore, we design the Global Context Flow (GCF) module to generate the global context information at different stages, which aims to learn the relationship among different salient regions and alleviate the dilution effect of high-level features. Experimental results on six benchmark datasets demonstrate that the proposed approach outperforms the state-of-the-art methods both quantitatively and qualitatively. 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

Study on the degree distribution properties of soil crack 3

Study on the degree distribution properties of soil crack

引用

3rd International Conference on Geoscience, Energy and Materials, I-GEM 2019

作者： Liu, Zuhan Jiangxi Province Key Laboratory for Water Information Cooperative Sensing and Intelligent Processing Nanchang Institute of Technology Nanchang Jiangxi China Key Laboratory of Poyang Lake Wetland and Watershed Research Ministry of Education Jiangxi Normal University Nanchang Jiangxi China Department of Geography and Geographic Information Science University of Cincinnati CincinnatiOH45221-0131 United States

The Dry-hot Valley Region (DVR) is one of the typical ecologically fragile areas, but also it is a key and difficult region of vegetation rehabilitation in southwestern mountain area of China. Due to the extremely hot dry climate and strong expansion/shrinkage characteristic of soil, the cracking is a kind of extremely universal phenomenon. Soil cracking accelerates the water loss by means of evaporation, intensifying the damage of soil drought to plants, so it has become is one of the important reasons for vegetation recovery. In this paper, according to complex network theory, we analyse the degree distribution properties of two typical soil-crack network in Yuanmou Dry-hot Valley Region, which includes not only the topological characteristics of the fracture network itself, but also more important study the reflected characteristics of the soil crack under different network topologies. © 2019 IOP Publishing Ltd. All rights reserved.

关键词： Soils

来源：评论

学校读者我要写书评

暂无评论

Spectral-Spatial Weighted Sparse Nonnegative Tensor Factorization for Hyperspectral Unmixing

Spectral-Spatial Weighted Sparse Nonnegative Tensor Factoriz...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Shaoquan Zhang Guorong Zhang Chengzhi Deng Jun Li Shengqian Wang Jun Wang Antonio Plaza Jiangxi Province Key Laboratory of Water Information Cooperative Sensing and Intelligent Processing Nanchang Institute of Technology Nanchang China Guangdong Provincial Key Laboratory of Urbanization and Geo-simulation School of Geography and Planning Sun Yat-sen University Guangzhou China Hyperspectral Computing Laboratory Escuela Politécnica University of Extremadura Cáceres Spain

ISBN: (数字)9781728163741

ISBN: (纸本)9781728163758

Hyperspectral unmixing aims to decompose a hyperspectral image (HSI) into a collection of constituent materials, or end-members, and their corresponding abundance fractions. Recently, nonnegative tensor factorization (NTF)-based spectral unmixing methods have attracted significant attention owing to their outstanding performance when representing an HSI without any information loss. However, tensor factorization-based HSI methods do not fully exploit the spatial contextual information present in the scene. Besides, these approaches are sensitive to low signal-to-noise ratio (SNR) in HSIs. To address this limitation, we propose a new spectral-spatial weighted sparse nonnegative tensor factorization (SSWNTF) method to preserve the spatial details in the abundance maps via the spectral and spatial weighting factors. Our experiments with simulated data sets certified that the proposed method outperforms other advanced methods.

关键词： Hyperspectral imaging Tensors Signal to noise ratio Signal processing algorithms Matrix decomposition Sparse matrices Noise measurement

来源：评论

学校读者我要写书评

暂无评论

Face Image Super-Resolution via Sparse Representation and Local Texture Constraints 15th

Face Image Super-Resolution via Sparse Representation and Lo...

引用

15th International Conference on intelligent Computing, ICIC 2019

作者： Wang, Bao-Cheng Li, Bo School of Computer Science of Technology Wuhan University of Science of Technology WuhanHubei430065 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System WuhanHubei430065 China

ISBN: (纸本)9783030267629

In order to solve the problem brought by low-resolution images and to make better use of low-resolution images, this paper proposed a face super-resolution image reconstruction method based on sparse representation and local texture constraints. In the proposed method, sparse representation is used to reconstruct the super-resolution of face image. Moreover, the local texture of the reconstructed image is also used twice before and after reconstruction as constraints to obtain better details and make the reconstructed face image clearer. Experimental results show that this method can effectively reconstruct the corresponding high-resolution images from low-resolution images. © 2019, Springer Nature Switzerland AG.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Relapse risk prediction for children with henoch-schonlein purpura based on GA-SVM

引用

International Journal Bioautomation 2020年第2期24卷 117-130页

作者： Liu, Yijun Wang, Beihong Li, Renpu He, Sheng Xi, Haixu Luo, Ye Key Laboratory of Cloud Computing and Intelligent Information Processing of Changzhou City Jiangsu University of Technology 1801 Zhongwu Road Changzhou213001 China Department of Pediatrics Changzhou No. 2 People's Hospital 29 Xinglong Road Changzhou213003 China

The relapse risk prediction for children with Henoch-Schonlein purpura can help pediatricians make an accurate prognosis and offer personalized and appropriate follow-up nursing and relapse control to patients. In this study, we propose a Genetic algorithm-Support vector machine (GA-SVM) learning method combining the support vector machine with the genetic algorithm for parameter optimization to capture the nonlinear mapping from a panel of biomarkers to the relapse risk of HSP children. The GA-SVM prediction model is created by using the dataset of 40 samples in clinical treatment and observation of patients. The inputs of the model consist of 19 biomarkers including gender, age, immunoglobulin M, immunoglobulin G, immunoglobulin A, prothrombin time, etc. The outputs consist of 1 and-1, where 1 indicates high relapse risk and-1 indicates low relapse risk. For comparison, the GS-SVM prediction model based on parameter optimization of grid search is also created. The experimental results show that the GA-SVM prediction model has a high prediction accuracy of 90% and is strong in generalization ability. The GA-SVM model for predicting the relapse risk of HSP children is a promising decision support tool of clinical prognosis, which provides pediatricians with valuable assistance to offer rehabilitation treatment to patients. © 2020 by the authors.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Graphene based functional devices: A short review

引用

Frontiers of physics 2019年第1期14卷 23-42页

作者： Rong Wang Xin-Gang Ren Ze Yan Li-Jun Jiang Wei E. I. Sha Guang-Cun Shan Department of Electrical and Electronic Engineering The University of Hong Kong Pokfulam Road Hong Kong China Key Laboratory of Intelligent Computing & Signal Processing Ministry of Education Anhui University Hefei 230039 China School of Instrumentation Science & Opto-electronics Engineering Beihang University Beijing 100191 China Key Laboratory of Micro-nano Electronic Devices and Smart Systems of Zhejiang Province College of Information Science & Electronic Engineering Zhejiang University Hangzhou 310027 China California NanoSystem Institute and Department of Chemistry & Biochemistry University of California Los Angeles CA 90095 USA

Graphene is an ideal 2D material system bridging electronic and photonic devices. It also breaks the fundamental speed and size limits by electronics and photonics, respectively. Graphene offers multiple functions of signal transmission, emission, modulation, and detection in a broad band, high speed, compact size, and low loss. Here, we have a brief view of graphene based functional devices at microwave, terahertz, and optical frequencies. Their fundamental physics and computational models were discussed as well.

关键词： graphene terahertz antenna microwave

来源：评论

学校读者我要写书评

暂无评论

RTM3D: Real-time monocular 3D detection from object keypoints for autonomous driving

arXiv

引用

arXiv 2020年

作者： Li, Peixuan Zhao, Huaici Liu, Pengfei Cao, Feidao Shenyang Institute of Automation Chinese Academy of Sciences Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences University of Chinese Academy of Sciences Key Laboratory of Opto-Electronic Information Processing Chinese Academy of Sciences Key Lab of Image Understanding and Computer Vision Liaoning Province

In this work, we propose an efficient and accurate monocular 3D detection framework in single shot. Most successful 3D detectors take the projection constraint from the 3D bounding box to the 2D box as an important component. Four edges of a 2D box provide only four constraints and the performance deteriorates dramatically with the small error of the 2D detector. Different from these approaches, our method predicts the nine perspective keypoints of a 3D bounding box in image space, and then utilize the geometric relationship of 3D and 2D perspectives to recover the dimension, location, and orientation in 3D space. In this method, the properties of the object can be predicted stably even when the estimation of keypoints is very noisy, which enables us to obtain fast detection speed with a small architecture. Training our method only uses the 3D properties of the object without the need for external networks or supervision data. Our method is the first real-time system for monocular image 3D detection while achieves state-ofthe-art performance on the KITTI benchmark. Code will be released at https://***/Banconxuan/RTM3D. Copyright © 2020, The Authors. All rights reserved.

关键词： Real time systems

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：