检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2025年第6期6卷 1488-1501页

作者： Saday, Artun Demirel, Ilker Yildirim, Yigit Tekin, Cem Bilkent University Department of Electrical and Electronics Engineering Ankara06800 Turkey Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory CambridgeMA02139 United States

Multi-armed bandits (MAB) is a sequential decision-making model in which the learner controls the trade-off between exploration and exploitation to maximize its cumulative reward. Federated multi-armed bandits (FMAB) is an emerging framework where a cohort of learners with heterogeneous local models play a MAB game and communicate their aggregated feedback to a server to learn a globally optimal arm. Two key hurdles in FMAB are communication-efficient learning and resilience to adversarial attacks. To address these issues, we study the FMAB problem in the presence of Byzantine clients who can send false model updates threatening the learning process. We analyze the sample complexity and the regret of β-optimal arm identification. We borrow tools from robust statistics and propose a median-of-means (MoM)-based online algorithm, Fed-MoM-UCB, to cope with Byzantine clients. In particular, we show that if the Byzantine clients constitute less than half of the cohort, the cumulative regret with respect to β-optimal arms is bounded over time with high probability, showcasing both communication efficiency and Byzantine resilience. We analyze the interplay between the algorithm parameters, a discernibility margin, regret, communication cost, and the arms' suboptimality gaps. We demonstrate Fed-MoM-UCB's effectiveness against the baselines in the presence of Byzantine attacks via experiments. © 2020 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Ensemble-based Machine Learning Classification for Early Detection of Autism Spectrum Disorder 3

Ensemble-based Machine Learning Classification for Early Det...

引用

3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025

作者： Subhash, Ambika Rani Ashwin Kumar, U.M. School of Computer Science and Engineering REVA University Bengaluru India Department of Information Science and Engineering BMS Institute of Technology and Management Bengaluru India

ISBN: (纸本)9798331527549

A complicated neurological illness, autism spectrum disorder (ASD) impacts social skills, communication, and other behavioral elements. Because the symptoms of ASD might be confused with those of other mental health conditions, diagnosing it can take a long time. The goal of this study is to develop a categorization model that can forecast the risk of ASD in toddlers, kids, teens, and adults. To improve speed, we used a variety of machine learning methods and ensembled them. Four publicly accessible non-clinical ASD screening datasets from Kaggle and the UCI library were used to assess the models. These datasets included 1,054 cases with 19 features for toddlers and different characteristics for older age groups. The classifiers' accuracy was almost 99%, according to the results. © 2025 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

X-News dataset for online news categorization

引用

International Journal of Intelligent Computing and Cybernetics 2024年第4期17卷 737-758页

作者： Samia Nawaz Yousafzai Hooria Shahbaz Armughan Ali Amreen Qamar Inzamam Mashood Nasir Sara Tehsin Robertas Damasevicius Department of Computer Science HITEC UniversityTaxila CanttPakistan Department of Statistics Quaid-I-Azam UniversityIslamabadPakistan Centre of Real Time Computer Systems Kaunas University of TechnologyKaunasLithuania Vytautas Magnus University KaunasLithuania

Purpose-The objective is to develop a more effective model that simplifies and accelerates the news classification process using advanced text mining and deep learning(DL)techniques.A distributed framework utilizing Bidirectional Encoder Representations from Transformers(BERT)was developed to classify news *** approach leverages various text mining and DL techniques on a distributed infrastructure,aiming to offer an alternative to traditional news classification ***/methodology/approach-This study focuses on the classification of distinct types of news by analyzing tweets from various news *** addresses the limitations of using benchmark datasets for news classification,which often result in models that are impractical for real-world ***-The framework’s effectiveness was evaluated on a newly proposed dataset and two additional benchmark datasets from the Kaggle repository,assessing the performance of each text mining and classification method across these *** results of this study demonstrate that the proposed strategy significantly outperforms other approaches in terms of accuracy and execution *** indicates that the distributed framework,coupled with the use of BERT for text analysis,provides a robust solution for analyzing large volumes of data *** findings also highlight the value of the newly released corpus for further research in news classification and emotion classification,suggesting its potential to facilitate advancements in these ***/value-This research introduces an innovative distributed framework for news classification that addresses the shortcomings of models trained on benchmark *** utilizing cutting-edge techniques and a novel dataset,the study offers significant improvements in accuracy and processing *** release of the corpus represents a valuable contribution to the field,enabling further exploration into news and emotion ***

关键词： News categorization BERT classifier POS tagging Social media analytics Deep learning for text analysis Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

CASA: Causality-driven Argument Sufficiency Assessment

CASA: Causality-driven Argument Sufficiency Assessment

引用

2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2024

作者： Liu, Xiao Feng, Yansong Chang, Kai-Wei Wangxuan Institute of Computer Technology Peking University China Computer Science Department University of California Los Angeles United States

ISBN: (纸本)9798891761148

The argument sufficiency assessment task aims to determine if the premises of a given argument support its conclusion. To tackle this task, existing works often train a classifier on data annotated by humans. However, annotating data is laborious, and annotations are often inconsistent due to subjective criteria. Motivated by the definition of probability of sufficiency (PS) in the causal literature, we propose CASA, a zero-shot causality-driven argument sufficiency assessment framework. PS measures how likely introducing the premise event would lead to the conclusion when both the premise and conclusion events are absent. To estimate this probability, we propose to use large language models (LLMs) to generate contexts that are inconsistent with the premise and conclusion and revise them by injecting the premise event. Experiments on two logical fallacy detection datasets demonstrate that CASA accurately identifies insufficient arguments. We further deploy CASA in a writing assistance application, and find that suggestions generated by CASA enhance the sufficiency of student-written arguments. Code and data are available at https://***/xxxiaol/CASA. © 2024 Association for Computational Linguistics.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

SDN-DDoSNet: A Deep Learning Framework for DDoS Attack Detection in Software-Defined Networks 8

SDN-DDoSNet: A Deep Learning Framework for DDoS Attack Detec...

引用

8th IEEE International Conference on Information and Communication technology, CICT 2024

作者： Ojha, Abhilash Yadav, Ashok Singh, Vrijendra Tezpur University Department of Computer Science Assam Tezpur India Indian Institute of Information Technology Department of Information Technology Allahabad Prayagraj India

ISBN: (纸本)9798331531539

Distributed Denial of Service (DDoS) attack is a malicious attempt to disrupt normal web traffic, posing an increasing threat to network infrastructures due to the growing frequency and sophistication of such attacks. This necessitates the development of robust and adaptive defense mechanisms. In this research, we propose a deep learning-based approach for DDoS attack detection, leveraging a Software-Defined Net-working dataset. Our study explores three models: Long-short-term memory,convolutional neural network, and hybrid CNN-LSTM architecture, each designed to capture the temporal and spatial features inherent in network traffic data. The data set includes diverse attack scenarios, including UDP, ICMP, and SYN protocol. The LSTM model achieved an accuracy of 99. 61% in the SDN- DDOS data set, followed by the hybrid model at 99.24%, and the CNN model at 98.86%. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Contribution Unveiling Cutting-Edge Machine Learning Techniques for Image Segmentation

Contribution Unveiling Cutting-Edge Machine Learning Techniq...

引用

International Conference on Electrical and Electronics Engineering, ICEEE 2023

作者： Shaik, Nazeer Gupta, Ankur Bhati, Sunita Kumar, Jaideep Singh, Jagendra Budhiraja, Ishan Department of Computer Science and Engineering Srinivasa Ramanujan Institute of Technology Andhra Pradesh Anantapur India School of Computer Science Engineering and Technology Bennett University Greater Noida India Department of Computer Applications JECRC University Jaipur India Raj Kumar Goel Institute of Technology AKTU University Lucknow India

ISBN: (纸本)9789819986606

Segregation of images is a critical step in processing images, computer vision, and a variety of other disciplines. The technique involves decomposing an illustration into numerous components or components, every single one that consists of an ensemble of elements with identical features for example African descent, frequency, or consistency. The most important objective of appearance, the intention of fragmentation seems to reduce complexity or customize the mathematical representation of an illustration in a way that is more readily reasonable and more straightforward for assessment. It is widely employed to locate boundaries and features in photos, and this is favorable in an assortment of industries including clinical imaging, object detection, recognition, and autonomous vehicles. Several image segmentation techniques are available, and in incorporating thresholding, zone is an area-based differentiation and corner-based recognition. A threshold segment is a simple and commonly used technique that involves setting a threshold value and dividing the pixels into two classes based on their intensity values. Region-based segmentation involves grouping pixels based on their spatial proximity and similarity in characteristics, while edge-based segmentation involves detecting edges or boundaries in an image and using them to separate different regions. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Maximizing (k, L)-Core With Edge Augmentation in Multilayer Graphs

引用

IEEE Transactions on Computational Social Systems 2024年第3期11卷 3931-3943页

作者： Chang, Chih-Chieh Lu, Chia-Hsun Teng, Shun-Jen Chang, Ming-Yi Ho, Ya-Chi Shen, Chih-Ya National Taiwan University of Science and Technology Management Undergraduate Program Taipei106 Taiwan National Tsing Hua University Department of Computer Science Hsinchu300 Taiwan Fu Jen Catholic University Department of Sociology Taipei242 Taiwan Hsinchu310 Taiwan

While most previous work pays attention on extracting dense subgraphs, such as k-cores, we argue that augmenting the graph to maximize the size of dense subgraphs is also very important and finds many applications. Therefore, in this article, we study the dense subgraph augmentation problem in multilayer graphs. Specifically, we propose the notion of (k, L)-core to model the dense subgraphs in multilayer graphs and propose a new research problem, budgeted maximal (k, L)-core augmentation (BMA) problem, which adds at most b edges in the multilayer graphs to maximize the size of (k, L)-core. We prove the NP-hardness of the general BMA problem when k = 2 and devise a polynomial-time algorithm to find the optimal solution for a special case of BMA, i.e., (2,1)-BMA. We then devise an effective algorithm, named search for optimum and reorder adaptively (SORA), with various performance-improving strategies to tackle the general BMA problem. We evaluate the performance of the proposed approaches on multiple large-scale datasets and compare them with the state-of-the-art baselines. Experimental results indicate that our proposed approaches significantly outperform the baselines in terms of solution quality and efficiency. © 2014 IEEE.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Multimodal Brain Tumor Segmentation Using Transformer and UNET 20

Multimodal Brain Tumor Segmentation Using Transformer and UN...

引用

20th International computer Conference on Wavelet Active Media technology and Information Processing, ICCWAMTIP 2023

作者： Osei, Isaac Appiah-Kubi, Benjamin Frimpong, Bill Kissi Hayfron-Acquah, James Ben Owusu-Agyemang, Kwabena Ofori-Addo, Samuel Turkson, Regina Esi Mawuli, Cobbinah Bernard Kwame Nkrumah University of Science and Technology Department of Computer Science Kumasi Ghana Ho Technical University Department of Computer Science Ho Ghana University of Cape Coast PMB University Post Office Department of Computer Science and Information Technology Cape Coast Ghana School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China College of Management Information Systems University of Ghana Ghana

ISBN: (纸本)9798350318982

In medical image analysis, accurate 3D brain tissue segmentation is crucial. This study undertakes a meticulous comparative evaluation of two 3D segmentation models: the established 3DUnet and the innovative 3DUnet Transformer. The assessment is anchored on the segmentation accuracy of three critical brain tissue classes;Whole Tumor (WT) Enhancing Tumor (ET) and Tumor Core (TC), with the Dice similarity coefficient (DSC) serving as the principal metric of performance. 3DUnet model demonstrated a consistent advantage over the 3DUnet Transformer, yielding superior DSC scores for ET (0.771 vs. 0.753), WT (0.891 vs. 0.874), and TC (0.804 vs. 0.779). Complementing the DSC, the Hausdorff distance was employed as an additional metric, further evidencing the 3DUnet's enhanced precision in segmentation with lower distance values across the board. These findings highlight the 3DUnet's robust performance and its potential to significantly refine brain tissue segmentation processes within medical imaging applications. © 2023 IEEE.

关键词： Tumors

来源：评论

学校读者我要写书评

暂无评论

Enhanced CNN-SVM Fusion for In-Depth Facial Acne Analysis 3

Enhanced CNN-SVM Fusion for In-Depth Facial Acne Analysis

引用

3rd IEEE World Conference on Applied Intelligence and Computing, AIC 2024

作者： Prabha, Chander Monga, Sakshra Srivastava, Prakash Singla, Gagan Saluja, Nitin Kumar, Vinod Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Department of Computer Science & Engineering Dehradun India Chandigarh University Department of Computer Science & Engineering Punjab Mohali India

ISBN: (纸本)9798350384598

The purpose of this work is to present and access a comprehensive analysis of an advanced facial acne classification method that is focused on machine learning. Hormonal acne, pustules, papules, cysts, blackheads (open comedones), and whiteheads (closed comedones) were all included in the dataset that was used to train and evaluate the algorithm. Accuracy, recall, and F1-Score provide evidence that the model correctly recognizes a variety of acne types, with scores ranging from 93.98% to 95.86%. Other evaluation metrics also suggest that the model displays this accuracy and efficiency. As an additional point of interest, the support percentage demonstrates how samples are spread within each acne class, hence demonstrating the balanced distribution of classes within the dataset. The model demonstrated that it was very accurate, obtaining a total score of 98% for its performance. In addition, aggregate performance metrics such as micro, weighted, and macro averages provide further evidence that the algorithm is capable of providing a diagnosis of acne that is both accurate and trustworthy across all types of acne. The research contributes to the development of automated systems for detecting acne on the face, and dermatologists may find it beneficial to utilize the data acquired to improve their ability to diagnose and treat patients who suffer from facial acne and health issues. © 2024 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

New Optimization Method Based on Binary Tumbleweed Algorithm 1

引用

18th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2022

作者： Pan, Jeng-Shyang Wang, Gui-Ling Yang, Dongqiang Chu, Shu-Chuan College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China School of Computer Science and Technology Shandong Jianzhu University Jinan China Department of Information Management Chaoyang University of Technology Taichung41349 Taiwan

ISBN: (数字)9789819906055

ISBN: (纸本)9789819906048

Tumbleweed algorithm (TA) is a meta-heuristic algorithm that simulates the growth of Tumbleweed. Compared with other advanced meta-heuristic algorithms, TA has strong optimization performance and competitiveness. However, TA is a continuous optimization algorithm, which is not suitable for binary problems. This paper presents a binary Tumbleweed optimization algorithm (BTA). The continuous search space is mapped into the discrete space by V-shaped and S-shaped transfer functions. In addition, this paper uses the CEC2013 benchmark function to evaluate the efficiency of BTA. To prove the effectiveness of the proposed algorithm, the BTA is compared with four advanced binary optimization algorithms. The experimental results reflect that BTA has preferable optimization capabilities on unimodal, multimodal, and composite functions. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Heuristic algorithms

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：