检索结果-内蒙古大学图书馆

Multi-objective evolutionary optimization for hardware-aware neural network pruning

Fundamental Research 2024年第4期4卷 941-950页

作者： Wenjing Hong Guiying Li Shengcai Liu Peng Yang Ke Tang Guangdong Provincial Key Laboratory of Brain-Inspired Inteligent Computation Department of Computer Science and EngineeringSouthern University of Science and TechnologyShenzhen 518055China Research Institute of Trustworthy Autonomous Systems Southern University of Science and TechnologyShenzhen 518055China Department of Statistics and Data Science Southern University of Science and TechnologyShenzhen 518055China

Neural network pruning is a popular approach to reducing the computational complexity of deep neural *** recent years,as growing evidence shows that conventional network pruning methods employ inappropriate proxy metrics,and as new types of hardware become increasingly available,hardware-aware network pruning that incorporates hardware characteristics in the loop of network pruning has gained growing attention,Both network accuracy and hardware efficiency(latency,memory consumption,etc.)are critical objectives to the success of network pruning,but the conflict between the multiple objectives makes it impossible to find a single optimal *** studies mostly convert the hardware-aware network pruning to optimization problems with a single *** this paper,we propose to solve the hardware-aware network pruning problem with Multi-Objective Evolutionary Algorithms(MOEAs).Specifically,we formulate the problem as a multi-objective optimization problem,and propose a novel memetic MOEA,namely HAMP,that combines an efficient portfoliobased selection and a surrogate-assisted local search,to solve *** studies demonstrate the potential of MOEAs in providing simultaneously a set of alternative solutions and the superiority of HAMP compared to the state-of-the-art hardware-aware network pruning method.

关键词： Multi-objective optimization Evolutionary algorithm Neural network pruning Hardware-awaremachine learning Hardware efficiency

来源：评论

学校读者我要写书评

暂无评论

A Deterministic Search Approach for Solving Stochastic Drone Search and Rescue Planning Without Communications 17th

A Deterministic Search Approach for Solving Stochastic Drone...

引用

17th International Symposium on Combinatorial Search, SoCS 2024

作者： Mishlyakov, Evgeny Gruntov, Mikhail Shleyfman, Alexander Karpas, Erez Faculty of Data and Decision Sciences Technion Israel Department of Computer Science Bar-Ilan University Israel

ISBN: (纸本)9781577358916

In disaster relief efforts, delivering aid to areas with no communication poses a significant challenge. Unmanned aerial vehicles (UAVs) can be utilized to deliver aid kits to survivors in hard-to-reach areas;unfortunately, in some areas, lack of communication and infrastructure presents a key problem. In this paper, we address a stochastic planning problem of planning for a set of UAVs that deliver aid kits to areas that lack communications, where we do not know in advance the locations where aid kits need to be delivered, but rather have probabilistic information about the locations of aid targets. Our main insight is that, despite the stochastic nature of this problem, we can solve it through deterministic search by monitoring the expected reward for each partial solution. This insight enables the application of deterministic planning techniques, empirically demonstrating a notable improvement in efficiency and response speed. Our approach presents a promising solution to addressing the challenge of delivering aid in regions with limited radio infrastructure, as well as similar planning problems. © 2024, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Using analytic hierarchy process to evaluate deep learning for infrared target recognition

引用

Multimedia Tools and Applications 2024年第38期83卷 86229-86245页

作者： Xu, Laixiang Cao, Bingxu Wei, Xinhong Li, Bei Su, Jingfeng Ning, Shiyuan Zhao, Junmin School of Computer and Data Science Henan University of Urban Construction Pingdingshan467036 China School of Information Engineering Luohe Vocational Technology College Henan University of Technology Luohe Institute of Technology Luohe462000 China School of Computer and Data Science Innovation Laboratory of Smart Traffic and Big Data Development of Hainan Province Pingdingshan467036 China School of Computer and Data Science Research Center of Smart City and Big Data Engineering of Henan Province Henan University of Urban Construction Pingdingshan467036 China Department of Software Information China Electronics Technology Group Corporation 36-th Research Institute Jiaxing314033 China School of Computer and Data Science Research Center of Smart City and Big Data Engineering of Henan Province Research Center of Intelligent Campus Application Engineering of Henan Province Innovation Laboratory of Smart Traffic and Big Data Development of Hainan Province Henan University of Urban Construction Pingdingshan467036 China

In order to promote the evaluation performance of deep learning infrared automatic target recognition (ATR) algorithms in the complex environment of air-to-air missile research, we proposed an analytic hierarchy process (AHP) evaluation system consisting of the target recognition ability, the generalization recognition ability, the anti-interference ability, the background suppression ability, and the cost metrics ability. According to detailed quantitative analysis and experimental results, we establish probability evaluation models and get their probability values. Compared with traditional evaluation systems, our proposed method implementation is showing high reliability. Similar superior performances are also achieved, such as quick, comprehensive, and scientific. Through the above research, the key technology of deep learning performance evaluation in automatic target recognition has been broken through, which provides some basis and guidance for the analysis of infrared imaging-guided weapon systems. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Infrared imaging

来源：评论

学校读者我要写书评

暂无评论

Contrastive Consensus Graph Learning for Multi-View Clustering

引用

IEEE/CAA Journal of Automatica Sinica 2022年第11期9卷 2027-2030页

作者： Shiping Wang Xincan Lin Zihan Fang Shide Du Guobao Xiao the College of Computer and Data Science Fuzhou UniversityFuzhou 350108China the College of Computer and Control Engineering Minjiang UniversityFuzhou 350108China

Dear Editor,This letter proposes a contrastive consensus graph learning model for multi-view *** are usually built to outline the correlation between multi-model objects in clustering task,and multiview graph clusteri... 详细信息

关键词： letter integrate property

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing Influenza A Detection in Humans: Unleashing the Power of XGBoost Classification 2

Revolutionizing Influenza A Detection in Humans: Unleashing ...

引用

2nd IEEE International Conference for Women in Innovation, Technology and Entrepreneurship, ICWITE 2024

作者： Nikhil, V. Sahaaya Arul Mary, S.A. Natarajan, B. Bajulunisha, A. Anuradha, R.S. Venkatraman, K. Chennai India Vellore Institute of Technology School of Computer Science and Engineering Vellore India VIT University School of Computer Science Engineering Chennai India M.A.M College of Engineering and Technology Department of Artificial Intelligence & Data Science Tiruchirappalli621105 India Panimalar Engineering College Department of Artificial Intelligence & Data Science Chennai India

ISBN: (纸本)9798350383287

Influenza A, a zoonotic virus potentially affecting and infecting humans, poses a significant global health threat. This research paper presents a comprehensive study on predicting Influenza A outbreaks by applying the XGBoost classification algorithm. The study utilizes two substantial datasets: one from the World Health Organization (WHO) containing global Influenza A epidemiological data from 2003 to 2023 and the other encompassing Influenza A infections in New York counties from 2009 to 2019. Both datasets underwent pre-processing, with essential features selected to develop an XGBoost Classification model for forecasting Influenza A cases in humans. The models exhibited impressive accuracy and performance metrics. Specifically, the WHO dataset demonstrated an AUC of 0.996 and an accuracy score of 0.99, while the New York dataset displayed an AUC of 0.992 and an accuracy score of 0.978. These outcomes underscore the model's capability to differentiate between positive and negative cases, showcasing high precision, accuracy, and F1 scores. The implications of this research extend to various domains, including the early detection of outbreaks, informed decision-making processes, and the implementation of public health interventions. The findings emphasize the pivotal role of high-quality and comprehensive datasets in training effective predictive models. Furthermore, the study points towards potential future avenues, such as personalized predictions, geo-spatial analysis, and integrating these predictive models with healthcare systems. © 2024 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

The Generalized Fixed Point Iteration Method for AVE

引用

Journal of Mathematical Research with Applications 2024年第6期44卷 837-849页

作者： Changfeng MA Jing KANG School of Data and Computer Science Xiamen Institute of Technology College of Mathematics and Informatics Fujian Normal University

In this paper, based on the previous published work by Ke et al.(2019) and Li et al.(2022), by using the matrix splitting technique, generalized fixed point iteration method(GFPI) is established to solve the absolute value equation(AVE). The proposed method not only includes SOR-like method, FPI method, MFPI method and so on, but also generates some special versions. Some convergence conditions of the proposed method with different iteration error norms are presented. Furthermore, methods corresponding to other splitting methods are studied in detail. The effectiveness and feasibility of the proposed method are confirmed by some numerical experiments.

关键词： absolute value equations fixed point iteration convergence

来源：评论

学校读者我要写书评

暂无评论

Enhancing Urdu Intrinsic Plagiarism Detection Through Stylometry Features and Machine Learning 25

Enhancing Urdu Intrinsic Plagiarism Detection Through Stylom...

引用

25th International Multi Topic Conference, INMIC 2023

作者： Manzoor, Muhammad Faraz Farooq, Muhammad Shoaib Haseeb, Muhammad Farooq, Uzma Abid, Adnan Saeed, Ahmer University of Management and Technology Department of Computer Science Lahore Pakistan University of the Punjab Department of Data Science Lahore Pakistan Aircod Technologies Department of Computer Science Lahore Pakistan

ISBN: (纸本)9798350317701

The creation of digital content and the easy accessibility of information have led to a surge in academic and textual plagiarism. Plagiarism detection in multiple languages is essential to maintain the integrity of academic and literary works. In the context of the Urdu language, there is a growing need for effective plagiarism detection methods that are tailored to its unique linguistic characteristics. Existing Urdu plagiarism detection tools often rely on external sources or lack robustness in handling intrinsic forms of plagiarism, where the copied content is slightly modified or paraphrased. This research aims to bridge this gap by developing an intrinsic plagiarism detection system for the Urdu language, using a combination of machine learning, ensemble learning and Multi-Layer Perceptron (MLP). Furthermore, to train and evaluate our plagiarism detection models, we manually curate a corpus comprising a substantial collection of 1807 documents in Urdu. This corpus forms the foundation of our research, enabling us to develop and fine-tune our detection algorithms to effectively identify instances of intrinsic plagiarism in Urdu text. To comprehensively assess the unique stylistic fingerprints of documents, we employ a diverse set of word based stylometry features. This multifaceted approach enhances our ability to pinpoint instances of plagiarism in a robust manner. This research contributes to the ongoing efforts to combat plagiarism and uphold the integrity of written content, particularly in the context of the Urdu language, while also showcasing the effectiveness of different word based stylometry features in addressing this critical issue. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Automating Robot Design with Multi-Level Evolution 13

Automating Robot Design with Multi-Level Evolution

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Nitschke, Geoff Howard, David Aslan, Bilal University of Cape Town Department of Computer Science South Africa Data61 Csiro Australia

ISBN: (纸本)9798350308365

In evolutionary robotics, Multi-Level Evolution (MLE) has been demonstrated for effective robot designs using a bottom-up approach, first evolving which materials to use for modular components and then how these components are connected into a functional robot design. This paper evaluates MLE robotic design, as an evolutionary design method on various task (robot ambulation) environments in comparison to human designed robots (pre-designed robot controller-morphology couplings). Results indicate that the MLE method evolves robots that are effective across increasingly difficult (locomotion) task environments, out-performing pre-designed robots, and thus provide further support for the efficacy of MLE as an evolutionary robotic design method. Furthermore, results indicate the MLE method enables the evolution of suitable robotic designs for various environments, where such designs would be non-intuitive and unlikely in conventional robotic design. © 2024 IEEE.

关键词： Machine design

来源：评论

学校读者我要写书评

暂无评论

Beyond Precision and Recall: Measuring Search Engine Consistency Using Rank Stability

引用

IEEE Access 2025年 13卷 92242-92259页

作者： Lakshmi, S. Thinakaran, Rajermani Batri, Krishnan Sowrirajan, R. Murugan, Sivaram Jain University Department of Electronics and Communication Engineering Karnataka Bengaluru India INTI International University Faculty of Data Science and Information Technology Negeri Sembilan Malaysia Jain University School of Computer Science and Engineering Karnataka Bengaluru India Dr. N.G.P. Arts and Science College Department of Mathematics Tamil Nadu Coimbatore India Sivas University of Science and Technology Faculty of Engineering and Natural Sciences Department of Computer Engineering Sivas58000 Turkey

Traditional information retrieval metrics such as precision, recall, and F-measure assess document relevance but fail to capture the stability of search rankings. Search engines frequently update their algorithms, leading to variations in retrieved documents and their rankings across identical queries. This study introduces a novel rank correlation measure that quantifies search result consistency by analyzing rank fluctuations in repeated queries, serving as a proxy for temporally separated searches. The proposed approach distinguishes between overlapping and non-overlapping documents, providing a detailed measure of ranking stability. Empirical evaluations comparing Google and Bing reveal that Google exhibits greater rank fluctuations for non-overlapping documents, whereas both search engines maintain relatively stable rankings for overlapping documents. These findings highlight the importance of rank-based evaluation frameworks in search engine analysis, extending beyond conventional retrieval metrics. By incorporating stability assessments, this research enhances the methodology for evaluating search consistency in dynamic web environments, contributing to process innovation in information retrieval (SDG 9: Industry, Innovation, and Infrastructure) and promoting fair, reliable access to information (SDG 16: Peace, Justice, and Strong Institutions). © 2013 IEEE.

关键词： Online searching

来源：评论

学校读者我要写书评

暂无评论

A Real-Time Framework for Automatic Sarcasm Detection Using Proposed Tensor-DNN-50 Algorithm 4th

A Real-Time Framework for Automatic Sarcasm Detection Using ...

引用

4th International Conference on Frontiers in Computing and Systems, COMSYS 2023

作者： Murthy, Jamuna S. Siddesh, G.M. Department of Computer Science and Engineering Ramaiah Institute of Technology Bengaluru India Department of Artificial Intelligence and Data Science Ramaiah Institute of Technology Bengaluru India

ISBN: (纸本)9789819726103

Social media platforms such as Twitter, Facebook, and Instagram are vast repositories of trending global news. They generate an enormous amount of data, offering a valuable resource for both academic researchers and IT marketing teams. These data can be harnessed for various insights and applications. A prominent and pressing research area within this data is the detection of sarcasm and fake news. The internet is rife with misleading and false information, often causing widespread confusion. Current research primarily categorizes news as either real or fake, using machine learning algorithms. However, this binary approach struggles with news that is partially true yet contains elements that could be deemed fake. To address this issue, our work introduces a real-time framework for sarcasm detection, utilizing an innovative deep learning technique called Tensor-DNN-50 for multiclass classification. The TDNN-50 model not only classifies news articles based on authenticity but also quantifies the level of sarcasm within them. Comparative analysis against leading techniques like SVM, att-RNN, and EANN demonstrates the superiority of our proposed algorithm. It achieves an impressive accuracy of 96.4% for two-class classification, 94.32% for four-class classification, and an astonishing 99.1% for seven-class classification. These results underscore the potential of our Tensor-DNN-50 Model in effectively addressing the intricate challenges of sarcasm detection and news classification with unparalleled accuracy across multiple classes. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Support vector machines

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：