检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2023年第1期4卷 107-120页

作者： Feng, Liang Shang, Qingxia Hou, Yaqing Tan, Kay Chen Ong, Yew-Soon College of Computer Science Chongqing University Chongqing400044 China College of Computer Science and Technology Dalian University of Technology Dalian116024 China Department of Computing The Hong Kong Polytechnic University Hong Kong Hong Kong Centre for Frontier AI Research A STAR Singapore138632 Singapore School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore

Large-scale optimization is vital in today's artificial intelligence (ai) applications for extracting essential knowledge from huge volumes of data. In recent years, to improve the evolutionary algorithms used to solve optimization problems involving a large number of decision variables, many attempts have been made to simplify the problem solution space of a given problem for the evolutionary search. In the literature, the existing approaches can generally be categorized as decomposition-based methods and dimension-reduction-based methods. The former decomposes a large-scale problem into several smaller subproblems, while the latter transforms the original high-dimensional solution space into a low-dimensional space. However, it is worth noting that a given large-scale optimization problem may not always be decomposable, and it is also difficult to guarantee that the global optimum of the original problem is preserved in the reduced low-dimensional problem space. This article, thus, proposes a new search paradigm, namely the multispace evolutionary search, to enhance the existing evolutionary search methods for solving large-scale optimization problems. In contrast to existing approaches that perform an evolutionary search in a single search space, the proposed paradigm is designed to conduct a search in multiple solution spaces that are derived from the given problem, each possessing a unique landscape. The proposed paradigm makes no assumptions about the large-scale optimization problem of interest, such as that the problem is decomposable or that a certain relationship exists among the decision variables. To verify the efficacy of the proposed paradigm, comprehensive empirical studies in comparison to five state-of-the-art algorithms were conducted using the CEC2013 large-scale benchmark problems as well as an ai application in e-commerce, i.e., movie recommendation. © 2022 IEEE.

关键词： Evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

Eyecheck Diabscan: A Productive Approach for Few-Shot Excellence in Retinopathy Diagnosis

Eyecheck Diabscan: A Productive Approach for Few-Shot Excell...

引用

Advanced Computing & Communication Technologies (ICACCTech), International Conference on

作者： Nuthanakanti Bhaskar E. Krishnaveni Sinde Jayavanth Rao Kumbala Pradeep Reddy G Vinoda Reddy Avala Raji Reddy Department of Computer Science and Engineering CMR Technical Campus Hyderabad India Department of Computer Science and Engineering CMR College of Engineering & Technology Hyderabad India Department of Computer Science and Engineering (DS) CMR Engineering College Hyderabad India Department of Computer Science and Engineering CMR Institute of Technology Hyderabad India Department of Computer Science and Engineering (AI&ML) CMR Technical Campus Hyderabad India Department of Mechanical Engineering CMR Technical Campus Hyderabad India

ISBN: (数字)9798331519056

ISBN: (纸本)9798331519063

The design and implementation of a few-shot learning-based diabetic retinopathy diagnostic system is the focus of this research project. According to the NEI (National Eye Institute), The United States report 2021, diabetic retinal degeneration is one of the most common causes of vision loss in people aged 20 to 74. Early recognition and treatment are therefore deemed essential. The National Eye Institute is a resource-based informational center about diseases and eye health. The primary objective of this endeavor is to improve the few-shot learning technique's capacity for early-stage diabetic eye disease detection. This suggested model may operate using a restricted amount of data, i.e., less data processing by enhancing data optimizing, and localize the necessary discriminative regions. Additionally, it enables the model to recognize and follow unique, novel patterns among the inputs. This approach is going to operate along with an assessment of proposals pertaining to patients' geographic and financial obstacles. Diabetic Retinopathy Identification is applied in the medical and healthcare sectors. This enables us to manage progression risk, forecast therapy results, and lessen adverse consequences and acute implications.

关键词： Diabetic retinopathy Accuracy Shape Shape measurement Retina Pattern recognition Ophthalmology Proposals Few shot learning Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Improving Gloss-free Sign Language Translation by Reducing Representation Density 38

Improving Gloss-free Sign Language Translation by Reducing R...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Ye, Jinhui Wang, Xing Jiao, Wenxiang Liang, Junwei Xiong, Hui Guangzhou China Department of Computer Science and Engineering HKUST Hong Kong Guangzhou HKUST Fok Ying Tung Research Institute Hong Kong Tencent AI Lab China

Gloss-free sign language translation (SLT) aims to develop well-performing SLT systems with no requirement for the costly gloss annotations, but currently still lags behind gloss-based approaches significantly. In this paper, we identify a representation density problem that could be a bottleneck in restricting the performance of gloss-free SLT. Specifically, the representation density problem describes that the visual representations of semantically distinct sign gestures tend to be closely packed together in feature space, which makes gloss-free methods struggle with distinguishing different sign gestures and suffer from a sharp performance drop. To address the representation density problem, we introduce a simple but effective contrastive learning strategy, namely SignCL, which encourages gloss-free models to learn more discriminative feature representation in a self-supervised manner. Our experiments demonstrate that the proposed SignCL can significantly reduce the representation density and improve performance across various translation frameworks. Specifically, SignCL achieves a significant improvement in BLEU score for the Sign Language Transformer and GFSLT-VLP on the CSL-Daily dataset by 39% and 46%, respectively, without any increase of model parameters. Compared to Sign2GPT, a state-of-the-art method based on large-scale pre-trained vision and language models, SignCL achieves better performance with only 35% of its parameters. Implementation and Checkpoints are available at https://***/JinhuiYE/SignCL. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Quasi-Oppositional-Based Learning Aquila Optimization Routing Protocol in Energy-Aware Wireless Communication

The Quasi-Oppositional-Based Learning Aquila Optimization Ro...

引用

International Conference on 6G Communications Networking and Signal Processing, SGCNSP 2023

作者： Haroon, P. S. Abdul Lateef Hussein, A.H.A. Yamsani, Nagendar Pareek, Piyush Kumar Sindgi, Anupama Department of ECE Ballari Institute of Technology and Management Ballari India College of Pharmacy Ahl Al Bayt University Karbala Iraq School of Computer Science and Artificial Intelligence SR University Warangal India Department of AI and ML Nitte Meenakshi Institute of Technologym Bengaluru India Department of ECE Presidency University Bengaluru India

ISBN: (纸本)9789819778751

Wireless Sensor Networks (WSN) is the extensive intelligent data system which combines the data collection, transmission as well as processing. In WSN, the routing becomes the crucial task that should controlled judiciously and the major objective of the routing approach is to transmit the data among Sensor Nodes (SN) as well as Base Stations (BS) to achieve communication. The energy consumption, scalability, deployment of the node are the difficulties in routing protocol. In this research, Quasi-Oppositional-based Learning Aquila Optimization (QOBL-AO) based routing protocol is proposed for the wireless communication. The objective of the proposed QOBL-AO approach is to achieve the Energy-Aware Routing (EAR) procedure of routing in Wireless Communication. The fitness function is obtained by utilizing the well-defined limitation named energy, delay, security as well as distance. The proposed QOBL-AO method attains the better results and it achieves an energy consumption of 0.35, PDR of 98.77%, End-To-End Delay (ETED) of 3.25 and network time of 5349 when compared to the existing methods such as Wavelet Mutation with AO-Based Energy Aware Routing (WMAO-EAR) and Metaheuristics Cluster-based Routing Technique for Energy-Efficient WSN (MHCRT-EEWSN). © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

Machine learning tool for exploring sentiment analysis on twitter data

引用

Materials Today: Proceedings 2022年 56卷 1927-1934页

作者： Biradar, Shanta H. Gorabal, J.V. Gupta, Gaurav Department of Information Science & Engineering Sir M Visvesvaraya Institute of Technology Bengaluru562157 India Department of Computer Science & Engineering ATME College of Engineering Mysore570028 India Yogananda School of AI Computing and Data Science Shoolini University Himachal Pradesh Solan India

In social media, micro-blogging is a common routine of most people around the world. Social media analytics gathering structure and unstructured Big-Data from various social sites and analyzing to make business decisions using Apache Hadoop and Apache Hive respectively. The objective of the research work was to develop big data technology used for gathering and handling large unstructured data from real time social media for sentiment analysis for identifying the brand and services. The methodology devised an algorithm based on sentimental analysis using customers review classification, which dealt with prepossessing the datasets, clustering of the data based on the specific domains, feature vector using n-gram models and tf-idf vectors extracts synonyms and classification sentiment analysis. The result shows that applied sentimental analysis with unsupervised clustering of data into specific domains and supervised machine learning techniques handle large amounts of twitter data in an efficient way. The developed tool 1.5 times faster than that of traditional database to Hadoop cluster and also the accuracy is nearly 80 %, which helps the user in computing, analyzing and interpreting interaction and associations between people, topics and ideas. © 2021

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

An Ensemble Approach of CNN and GRU Models for Network and Server Intrusion

An Ensemble Approach of CNN and GRU Models for Network and S...

引用

Emerging Technology (INCET), International Conference for

作者： Aniketh Reddy Adireddy Vishnu Kurnala Sindu Patlolla Gagandeep Arora Artificial Intelligence & Machine Learning Vardhaman College of Engineering Hyderabad India Computer Science & Engineering (AI&ML) Vardhaman College of Engineering Hyderabad India

ISBN: (数字)9798350361155

ISBN: (纸本)9798350361162

This research paper delves into the realm of cybersecurity by proposing an innovative ensemble model that blends Gated Recurrent Units (GRU) and Convolutional Neural Networks (CNN) for network and server intrusion detection. CNNs excel in spatial feature extraction, particularly suited for analyzing network traffic patterns, while GRUs are adept at capturing temporal dependencies within sequential data. The integration of these models aims to leverage their complementary strengths, enhancing the overall robustness of intrusion detection systems. Through comprehensive training and evaluation on diverse datasets, the ensemble model showcases promising results, demonstrating improved accuracy and resilience against adversarial attacks compared to standalone CNN and GRU models. In addition to making a significant contribution to the rapidly developing field of deep learning in cybersecurity, this research offers practical implications for the development of more secure network and server environments in the face of dynamic cyber threats.

关键词： Training Intrusion detection Telecommunication traffic Logic gates Feature extraction Robustness Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Biometric-Based Access Control Systems with Robust Facial Recognition in IoT Environments 3

Biometric-Based Access Control Systems with Robust Facial Re...

引用

3rd IEEE International Conference on Intelligent Techniques in Control, Optimization and Signal Processing, INCOS 2024

作者： Satish Babu, B.V. Manoranjini, J. Changala, Ravindra Aarif, Mohd Mary, S. Suma Christal Raj, I Infant Prasad v Potluri Siddhartha Institute of Technology Department of It Vijayawada India Chaitanya Bharathi Institute of Technology Department of Ai & Ml Hyderabad India Guru Nanak Institutions Technical Campus Department of Information Technology Hyderabad India Associate Editor-Publishing Global Research Network Noida India Panimalar Engineering College Department of Information Technology Poonamalle Chennai India K. Ramakrishnan College of Engineering Department of Computer Science and Engineering Tamil Nadu Trichy India

ISBN: (纸本)9798350361186

The integration of biometric-based access control systems with robust facial recognition technology within Internet of Things (IoT) environments holds paramount significance in addressing contemporary security challenges. The novelty of the research endeavours to contribute to this critical area by investigating and implementing a novel approach, employing Convolutional Neural Networks (CNN) for facial classification within the biometric framework. The significance of this study lies in its potential to enhance security measures in IoT environments, ensuring reliable and efficient access control. Previous attempts at integrating facial recognition into IoT have encountered notable challenges, including suboptimal accuracy, vulnerability to adversarial attacks, and scalability issues. This research builds upon these prior issues by introducing CNN for facial classification, leveraging its capabilities in image processing and pattern recognition to achieve enhanced accuracy and resilience against potential security threats. The novelty of the study is underscored by the application of CNN within the biometric context of IoT, offering a cutting-edge solution tailored to the intricacies of interconnected devices. Experiments conducted on benchmark datasets showcase the promising capabilities of the proposed system, with face recognition accuracies reaching an impressive 98%. These results underscore the potential of this research to significantly advance the state-of-the-art in biometric-based access control, particularly within the context of IoT environments. The utilization of CNN introduces a pioneering element, contributing to the evolution of secure and efficient access control solutions in the era of IoT. © 2024 IEEE.

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

An IoT-based Deep Belief Network in Intrusion Detection System using Butterfly Optimization 10

An IoT-based Deep Belief Network in Intrusion Detection Syst...

引用

10th International Conference on Communication and Signal Processing, ICCSP 2024

作者： Cherian, Aswathy K. Vaidhehi, M. Arshey, M. Simpson, Serin V. Briskilal, J. Singh, Astitva Srm Institute of Science and Technology Department of Computing Technologies Chennai Kattankulathur India Scms School of Engineering and Technology Department of Ai & Ds Kerala Ernakulam India Madanapalle Institute of Technology & Science Department of Computer Science and Technology Chittoor Andhra Pradesh Madanapalle India

ISBN: (纸本)9798350353068

Rapid technological advancements in the communication and internet industries have led to a huge increase in network scale and data volume. Network security is having a hard time identifying attacks as new threats are being created. Additionally, you can't overlook the possibility of unauthorized users trying to attack the network in various methods. Numerous technologies, such as an intrusion detection system, can be utilized to protect the network from potential threats (IDS). The collecting and dissemination of data by Internet-connected devices is referred to as the 'Internet of Things' (IoT). There are now more privacy and security problems due to the proliferation of Internet connections and the introduction of new technologies like the Internet of Things (IoT). IoT incursions are becoming increasingly complex and numerous as a result of the proliferation of connected devices. Companies are boosting their investment in research and development in order to better detect these threats. In this research, Deep Belief Network (DBN) and Butterfly Optimization Algorithm (BOA) Intrusion Detection Systems (IDS) are proposed for detecting Io T network assaults. In order to minimize information leakage on the test data, the UNSW-NB15 dataset was normalized using the Absolute Maximum Scaling (AMS) approach in the first stage of this study process. This dataset contains a mixture of current assaults and typical network traffic, categorized into nine different attack types. The next step was to use Principal Component Analysis to reduce the dimensionality (PCA). Last but not least, results showed that DBN-BOA was superior to other IoT intrusion detection systems in terms of accuracy and speed. © 2024 IEEE.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Network security, trust & privacy in a wiredwireless Environments-An Overview 1

Network security, trust & privacy in a wiredwireless Environ...

引用

1st International Conference on Computational Intelligence, Communication Technology and Networking, CICTN 2023

作者： Salama, Ramiz Al-Turjman, Fadi Bhatla, Shruti Gautam, Devendra Ai and Robotics Institute Research Center for Ai and IoT Near East University Nicosia Department of Computer Engineering Mersin10 Turkey Ai and Robotics Institute Near East University Artificial Intelligence Engineering Dept. Mersin Nicosia10 Turkey University of Kyrenia Research Center for Ai and IoT Faculty of Engineering Kyrenia Mersin10 Turkey Graphic Era Hill University Department of Computer Science and Engineering Dehradun India Abes Engineering College Ghaziabad India

ISBN: (纸本)9798350338027

In this report, we explore the key challenges and threats facing network security in wired and wireless environments, as well as the technologies and best practices for addressing these challenges. We provide an overview of the key concepts and terminology related to network security, trust, and privacy, and discuss the role of encryption in maintaining trust and privacy in these environments. We also examine common attack vectors and countermeasures and offer some recommendations for best practices in the realm of network security. © 2023 IEEE.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Cultural Alignment in Large Language Models: An Explanatory Analysis Based on Hofstede's Cultural Dimensions 31

Cultural Alignment in Large Language Models: An Explanatory ...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Masoud, Reem I. Liu, Ziquan Ferianc, Martin Treleaven, Philip Rodrigues, Miguel Department of Electronic and Electrical Engineering University College London United Kingdom Department of Computer Science University College London United Kingdom Department of Electrical Engineering King Abdulaziz University Saudi Arabia AI Centre University College London United Kingdom

ISBN: (纸本)9798891761964

The deployment of large language models (LLMs) raises concerns regarding their cultural misalignment and potential ramifications on individuals and societies with diverse cultural backgrounds. While the discourse has focused mainly on political and social biases, our research proposes a Cultural Alignment Test (Hoftede's CAT) to quantify cultural alignment using Hofstede's cultural dimension framework, which offers an explanatory cross-cultural comparison through the latent variable analysis. We apply our approach to quantitatively evaluate LLMs-namely Llama 2, GPT-3.5, and GPT-4-against the cultural dimensions of regions like the United States, China, and Arab countries, using different prompting styles and exploring the effects of language-specific fine-tuning on the models' behavioural tendencies and cultural values. Our results quantify the cultural alignment of LLMs and reveal the difference between LLMs in explanatory cultural dimensions. Our study demonstrates that while all LLMs struggle to grasp cultural values, GPT-4 shows a unique capability to adapt to cultural nuances, particularly in Chinese settings. However, it faces challenges with American and Arab cultures. The research also highlights that fine-tuning LLama 2 models with different languages changes their responses to cultural questions, emphasizing the need for culturally diverse development in ai for worldwide acceptance and ethical use. For more details or to contribute to this research, visit our GitHub page https://***/reemim/Hofstedes_CAT. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

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：