检索结果-内蒙古大学图书馆

Memristive Computing and Applications (ICMCA), IEEE International Conference on

作者： Aidi Liu Wenyue Yu Lili Huang Jian Guo Tengfei Lei Hui Zhao School of Computer and Information Engineering Qilu Institute of Technology Jinan China School of Computer and Information Engineering Qilu Institute of Technology Collaborative Innovation Center of Memristive Computing Application School of Computer and Information Engineering Qilu Institute of Technology Shandong Provincial Key Laboratory of Network Based Intelligent Computing School of Information Science and Engineering University of Jinan Jinan China

This article mainly conducts a series of research on the synchronization control of Cohen-Grossberg neural networks. Firstly, a memristive Cohen-Grossberg neural network model with time-varying delay was constructed, and an equivalent system was obtained by combining inverse function theory, differential inclusion, and set-valued mapping. Secondly, a new predefined-time stability theorem was proposed, and sufficient conditions were obtained to ensure that the drive-response system achieves predefined-time synchronization by designing appropriate control strategy. The simulation results show that the stability time of the error system can be adjusted by the user, verifying the correctness and of the theory.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Photon-Counting Integral Imaging Using Bayesian Estimation with Depth Camera Information

Photon-Counting Integral Imaging Using Bayesian Estimation w...

引用

International Conference on Control, Automation and Systems ( ICCAS)

作者： Hideaki Uchino Hyun-Woo Kim Myungjin Cho Min-Chul Lee Department of Computer Science and Networks Kyushu Institute of Technology Iizuka Fukuoka Japan School of ICT Robotics and Mechanical Engineering IITC Hankyong National University Anseong Kyonggi-do Republic of Korea

In this paper, we propose a method for visualizing object information under low light conditions using photon-counting integral imaging and depth images as prior information. To visualize 3D objects under low-light conditions, computational photon-counting imaging may be used. It is possible to estimate photons using statistical methods such as the Poisson distribution. Therefore, photon-counting integral imaging can visualize 3D images under low-light conditions by maximum likelihood estimation (MLE). However, MLE does not consider prior information so it may not be an accurate estimation. In addition, it is difficult to obtain accurate depth information as the distance increases due to rounding errors caused by computational reconstruction. Therefore, we propose Bayesian estimation such as a maximum a posteriori estimation method that uses as a priori information images from a depth camera, which can be acquired even in low-light conditions, to improve the quality of 3D images. To evaluate image quality of the proposed method, optical experiments were conducted by calculating image quality metrics. This technology can be utilized for night vision applications, such as object recognition in security camera systems and autonomous driving, using RGB cameras and depth cameras.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exploiting Long-term Temporal Dynamics for Video Captioning

arXiv

引用

arXiv 2022年

作者： Guo, Yuyu Zhang, Jingqiu Gao, Lianli The Future Media Center School of Computer Science and Engineering The University of Electronic Science and Technology of China China

Automatically describing videos with natural language is a fundamental challenge for computer vision and natural language processing. Recently, progress in this problem has been achieved through two steps: 1) employing 2-D and/or 3-D Convolutional Neural Networks (CNNs) (e.g. VGG, ResNet or C3D) to extract spatial and/or temporal features to encode video contents;and 2) applying Recurrent Neural Networks (RNNs) to generate sentences to describe events in videos. Temporal attention-based model has gained much progress by considering the importance of each video frame. However, for a long video, especially for a video which consists of a set of sub-events, we should discover and leverage the importance of each sub-shot instead of each frame. In this paper, we propose a novel approach, namely temporal and spatial LSTM (TS-LSTM), which systematically exploits spatial and temporal dynamics within video sequences. In TS-LSTM, a temporal pooling LSTM (TP-LSTM) is designed to incorporate both spatial and temporal information to extract long-term temporal dynamics within video sub-shots;and a stacked LSTM is introduced to generate a list of words to describe the video. Experimental results obtained in two public video captioning benchmarks indicate that our TS-LSTM outperforms the state-of-the-art methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Video recording

来源：评论

学校读者我要写书评

暂无评论

Enhancing Vegetable Quality Prediction with Fuzzy Interference System 2

Enhancing Vegetable Quality Prediction with Fuzzy Interferen...

引用

2nd International Conference on Business Analytics for Technology and Security, ICBATS 2023

作者： Al-Rajab, Murad Asif, Muhammad Chuhan, Saad Hussain Mustafa, Muzzamil Ilyas, Amna Kamran, Rukshanda Ahmad Abazeed, Ashraf Riad Abu Saima, Mahmoud Geeta, Sai Abu Dhabi University Computer Science and IT College of Engineering United Arab Emirates NCBA&E School of Computer Science Lahore Pakistan Applied Science Private University Applied Science Research Center Amman11937 Jordan Universiti Malaysia Faculty of Compter Science and Information Technology Sarawak94300 Malaysia ADVETI Abu Dhabi Vocational Education and Training Institute United Arab Emirates United Arab Emirates University College of Science United Arab Emirates Institute of Management Technology Dubai Head QAIE and Accreditation United Arab Emirates

ISBN: (纸本)9798350335644

We are living in this hurly burly world of Tumult and Turmoil where many problems arises who have not any exact solution/answer like we have health and nutrition problems people facing difficulties to select best vegetable for their health and if they select best vegetable for themselves they don't know about the quality of the vegetable which they have selects. In the vegetable processing industry, some manufacturers add extra ingredients to prolong the shelf life and maintain the quality of their products. However, excessive amounts of these ingredients can have negative health implications such as palpitations, headaches, allergies, and even cancer. Therefore, it is crucial to implement a system to assess the quality of vegetables being used, which can provide consumers and patients with information regarding their quality content. This system is particularly beneficial for addressing human-related issues, especially in determining the percentage of quality. Various methods have been established to achieve optimal solutions in response to rapidly changing living conditions. This paper proposes the development of a fuzzy-based system that takes inputs such as season, time, and condition to detect the quality of vegetables. The output of this system is determined using Kappa statistics. © 2023 IEEE.

关键词： Vegetables

来源：评论

学校读者我要写书评

暂无评论

Domain Adaptation in Visual Reinforcement Learning via Self-Expert Imitation with Purifying Latent Feature

Domain Adaptation in Visual Reinforcement Learning via Self-...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Lin Chen Jianan Huang Zhen Zhou Yaonan Wang Yang Mo Zhiqiang Miao Kai Zeng Mingtao Feng Danwei Wang School of Electrical and Information Engineering Hunan University Changsha China National Engineering Research Center for Robot Visual Perception and Control Technology Changsha China School of Computer Science and Technology Xidian University Xian China School of Electrical and Electrical Engineering Nanyang Technological University Singapore

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Generalizing visual reinforcement learning is fundamental to robot visual navigation, involving the acquisition of a policy from interactions with source environments to facilitate adaptation to analogous, yet unfamiliar target environments. Recent advancements capitalize on data augmentation techniques, self-supervised learning methods, and the generative adversarial network framework to train policy neural networks with enhanced generalizability. However, current methods, upon extracting domain-general latent features, further utilize these features to train the reinforcement learning policy, resulting in a decline in the performance of the learned policy guiding the agent to accomplish tasks. To tackle these challenges, a framework of self-expert imitation with purifying latent features was devised, empowering the policy to achieve robust and stable zero-shot generalization performance in visually similar domains previously unseen, without diminishing the performance of guiding the agent to accomplish tasks. The extraction method of domain-general latent features is proposed to enhance their quality based on the variational autoencoder. Extensive experiments have shown that our policy, compared with state-of-the-art counterparts, does not diminish the performance of the policy guiding the agent to accomplish tasks after generalization.

关键词： Training Visualization Navigation Semantic segmentation Autoencoders Neural networks Reinforcement learning Self-supervised learning Feature extraction Intelligent robots

来源：评论

学校读者我要写书评

暂无评论

Superluminal k-Gap Solitons in Nonlinear Photonic Time Crystals

引用

Physical Review Letters 2023年第23期130卷 233801-233801页

作者： Yiming Pan Moshe-Ishay Cohen Mordechai Segev Physics Department and Solid State Institute Technion Haifa 32000 Israel School of Physical Science and Technology and Center for Transformative Science ShanghaiTech University Shanghai 200031 China Department of Electrical and Computer Engineering Technion Haifa 32000 Israel

We propose superluminal solitons residing in the momentum gap (k gap) of nonlinear photonic time crystals. These gap solitons are structured as plane waves in space while being periodically self-reconstructing wave packets in time. The solitons emerge from modes with infinite group velocity causing superluminal evolution, which is the opposite of the stationary nature of the analogous Bragg gap soliton residing at the edge of an energy gap (or a spatial gap) with zero group velocity. We explore the faster-than-light pulsed propagation of these k-gap solitons in view of Einstein’s causality by introducing a truncated input seed as a precursor of a signal velocity forerunner, and find that the superluminal propagation of k-gap solitons does not break causality.

关键词： Light-matter interaction Metamaterials Nonlinear optics Photonic crystals Photonics Ultrafast optics

来源：评论

学校读者我要写书评

暂无评论

Intelligent Tutoring for Large-Scale Personalized Programming Learning Based on Knowledge Graph

Intelligent Tutoring for Large-Scale Personalized Programmin...

引用

Frontiers in Education (FIE) Conference

作者： Ying Li Jincheng Qiu Runze Yang Tongyu Zhu Hao Sheng ShiJie Gui Yu Liang School of Computer Science and Engineering Beihang University Beijing China Beijing Engineering Research Center for IoT Software and Systems Beijing University of Technology Beijing China

The development of learning analytics technology has created the necessary conditions for the implementation of large-scale personalized education. By taking advantage of the current advancements available, we can adopt a deep integration of “artificial intelligence + education”, take educational big data as the research object, aim at achieving precise teaching, and use cognitive diagnosis as a means to construct an intelligent guidance model. The model will conform to a learner's cognitive patterns and curriculum characteristics to meet his personalized learning needs under different knowledge states. Firstly, use knowledge map and abstract syntax trees to extract the characteristics of teaching objectives from questions and standard programing codes. Secondly, Use “target characteristics - learning characteristics - learning sequence - learning effect” cognitive diagnosis model to assess student achievement in learning. Finally, according to the correlation between knowledge points, the initial learning path consisting of all relevant knowledge points is identified from a knowledge map. And the most optimal learning path is selected based on the importance of knowledge points and student's learning situation. There are three innovation points in this paper. First, the traditional method which only focuses on discovering the one-dimensional knowledge features implicated in the questions, is extended to construct a knowledge-ability multidimensional feature matrix. Secondly, it optimized static cognitive diagnosis into dynamic cognitive arbitrariness, which has a modeling capability of dynamic temporal sequence. Thirdly, it regarded the correlation between knowledge points, and the importance of knowledge points as the heuristic conditions for constructing a learning path.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Innovation-Based Remote State Estimation Secrecy with no Acknowledgments

arXiv

引用

arXiv 2022年

作者： Kennedy, Justin M. Ford, Jason J. Quevedo, Daniel E. Dressler, Falko School of Electrical Engineering and Robotics Queensland University of Technology 2 George St BrisbaneQLD4000 Australia School of Electrical Engineering and Computer Science TU Berlin Germany

Secrecy encoding for remote state estimation in the presence of adversarial eavesdroppers is a well studied problem. Typical existing secrecy encoding schemes rely on the transmitter’s knowledge of the remote estimator’s current performance. This performance measure is often shared via packet receipt acknowledgments. However, in practical situations the acknowledgment channel may be susceptible to interference from an active adversary, resulting in the secrecy encoding scheme failing. Aiming to achieve a reliable state estimate for a legitimate estimator while ensuring secrecy, we propose a secrecy encoding scheme without the need for packet receipt acknowledgments. Our encoding scheme uses a pre-arranged scheduling sequence established at the transmitter and legitimate receiver. We transmit a packet containing either the state measurement or encoded information for the legitimate user. The encoding makes the packet appear to be the state but is designed to damage an eavesdropper’s estimate. The pre-arranged scheduling sequence and encoding is chosen psuedo-random. We analyze the performance of our encoding scheme against a class of eavesdropper, and show conditions to force the eavesdropper to have an unbounded estimation performance. Further, we provide a numerical illustration and apply our encoding scheme to an application in power systems. Copyright © 2022, The Authors. All rights reserved.

关键词： Transmitters

来源：评论

学校读者我要写书评

暂无评论

Hyperbolic Neural Network Based Preselection for Expensive Multi-Objective Optimization

引用

IEEE Transactions on Evolutionary Computation 2024年 1-1页

作者： Li, Bingdong Yang, Yanting Hong, Wenjing Yang, Peng Zhou, Aimin Shanghai Institute of AI for Education the School of Computer Science and Technology and the Shanghai Frontiers Science Center of Molecule Intelligent Syntheses East China Normal University Shanghai China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China Department of Computer Science and Engineering Guangdong Key Laboratory of Brain-Inspired Intelligent Computation Southern University of Science and Technology Shenzhen China

A series of surrogate-assisted evolutionary algorithms (SAEAs) have been proposed for expensive multi-objective optimization problems (EMOPs), building cheap surrogate models to replace the expensive real function evaluations. However, the search efficiency of these SAEAs is not yet satisfactory. More efforts are needed to further exploit useful information from the real function evaluations in order to better guide the search process. Facing this challenge, this paper proposes a Hyperbolic Neural Network (HNN) based preselection operator to accelerate the optimization process based on limited evaluated solutions. First, the preselection task is modeled as a multi-label classification problem where solutions are classified into different layers (ordinal categories) through -relaxed objective aggregation. Second, in order to resemble the hierarchical structure of candidate solutions, a hyperbolic neural network is applied to tackle the multi-label classification problem. The reason for using HNN is that hyperbolic spaces more closely resemble hierarchical structures than Euclidean spaces. Moreover, to alleviate the data deficiency issue, a data augmentation strategy is employed for training the HNN. In order to evaluate its performance, the proposed HNN-based preselection operator is embedded into two surrogate-assisted evolutionary algorithms. Experimental results on two benchmark test suites and three real-world problems with up to 11 objectives and 150 decision variables involving seven state-of-the-art algorithms demonstrate the effectiveness of the proposed method. IEEE

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Enhanced Analysis of Genomic Data for Precision Medicine

Machine Learning-Enhanced Analysis of Genomic Data for Preci...

引用

Emerging Trends in Industry 4.0 (ETI 4.0)

作者： R. Sasikala K. Jaya Deepthi T. Suresh Balakrishnan Prabhakar Krishnan U. Samson Ebenezar Department Of Computer Science And Engineering Care College Of Engineering Trichy Tamil Nadu Department Of Aiml Mohan Babu University Tirupati-2 Ap India Department Of Computer Science And Engineering Saveetha School Of Engineering Saveetha Institute Of Medical And Technical Sciences Saveetha University Chennai India Center For Cyber Security Systems And Networks Amrita Vishwa Vidyapeetham Amritapuri-Campus India Department Of Computer Science And Engineering Saveetha School Of Engineering Saveetha Institute Of Medical And Technical Sciences saveetha University Chennai Tamil Nadu India

ISBN: (数字)9798350373783

ISBN: (纸本)9798350373790

Precision medicine employs genetic data, which is enhanced with machine learning algorithms, to determine the most relevant treatment method per patient’s genotype. We have zoned in on the area of a machine learning-guided genetic information analysis in this study to assist in clinical decisionmaking of precision medicine while improvising. Genetic information for 500 cancer patients cases taken from the public database and cleaned to lean out the desired features were furnished. Different machine learning algorithms were trained, including SVM, random forest, gradient boost, deep learning (CNN) and ensemble (ensemble method). Their performance was investigated for predicting cancer subtypes. The ensemble model reached the top accuracy level ($88 \%$), the best sensitivity ($85 \%$), ever specificity ($91 \%$) and AUC score (0.95). By comparison a group had better result than an individual algorithm could across all metrics. In further, deep learning methods were reported to attain more accuracy and F1 scores which is more than traditional machine learning methods. These findings hence prequalify the usefulness of ML in investigating genotype data in precision medicine, and enable physicians with useful information to select treatment options, make prognosis, and indicate the medical care.

关键词： Deep learning Support vector machines Machine learning algorithms Accuracy Sensitivity Precision medicine Genomics

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：