检索结果-内蒙古大学图书馆

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Huang, Lifeng Su, Tian Gao, Chengying Liu, Ning Huang, Qiong College of Mathematics and Informatics South China Agricultural University China School of Computer Science and Engineering Sun Yat-sen University China

ISBN: (纸本)157735897X

Adversarial attacks poses a significant threat to the security of AI-based systems. To counteract these attacks, adversarial training (AT) and ensemble learning (EL) have emerged as widely adopted methods for enhancing model robustness. However, a counter-intuitive phenomenon arises where the simple combination of these approaches may potentially compromising adversarial robustness of ensemble models. In this paper, we propose a novel method called Alignment and Unlearning for Training Ensembles (AUTE), aiming to effectively integrate AT and EL to maximize their benefits. Specifically, AUTE incorporates two key components. Firstly, AUTE divides the ensemble into a big peer model and a single member in a loop manner, aligning their outputs for boosting robustness of each member. Secondly, AUTE introduces the concept of unlearning, actively forgetting specific data with over-confident properties to preserve model capacity to learn more robust features. Extensive experiments across various datasets and networks illustrate that AUTE achieves superior performance compared to baselines. For instance, a 5-member AUTE with ResNet-20 networks outperforms state-of-the-art method by 2.1% and 3.2% in classifying clean and adversarial data. Additionally, AUTE can easily extend to non-adversarial training paradigm, surpassing current standard ensemble learning methods by a large margin. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

QuarantivityVR: Supporting Self-Embodiment for Non-HMD Users in Asymmetric Social VR Games

引用

i-com 2022年第1期21卷 55-70页

作者： Yassien, Amal Soliman, Mohamed Ahmed Abdennadher, Slim German International University Department of Informatics and Computer Science Cairo Egypt German University in Cairo Department of Computer Science and Engineering Cairo Egypt

The prevalence of immersive head-mounted display (HMD) social virtual reality (VR) applications introduced asymmetric interaction among users within the virtual environment (VE). Therefore, researchers opted for (1) exploring the asymmetric social VR interaction dynamics in only co-located setups, (2) assigning interdependent roles to both HMD and non-HMD users, and (3) representing non-HMD users as abstract avatars in the VE. Therefore, we investigate the feasibility of supporting Self-Embodiment in an asymmetric VR interaction mode in a remote setup. To this end, we designed an asymmetric social VR game, QuarantivityVR, to (1) support sense of self-embodiment for non-HMD users in a remote setting by representing them as realistic full-body avatars within the VE, (2) augment visual-motor synchrony for the non-HMD users to increase their sense of agency and presence by detecting their motion through Kinect sensor and laptop's webcam. During the game, each player performs three activities in succession, namely movie-guessing, spelling-bee, and answering mathematical questions. We believe that our work will act as a step towards the inclusion of a wide spectrum of users that can not afford full immersion and will aid researchers in creating enjoyable interactions for both users in the physical and virtual spaces. © 2022 Walter de Gruyter GmbH, Berlin/Boston.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Optimizing the Clinical Decision Support System (CDSS) by Using Recurrent Neural Network (RNN) Language Models for Real-Time Medical Query Processing

引用

computers, Materials & Continua 2024年第12期81卷 4787-4832页

作者： Israa Ibraheem Al Barazanchi Wahidah Hashim Reema Thabit Mashary Nawwaf Alrasheedy Abeer Aljohan Jongwoon Park Byoungchol Chang College of Computing and Informatics Universiti Tenaga Nasional(UNITEN)Kajang43000Malaysia College of Engineering University ofWarith Al-AnbiyaaKarbala56001Iraq Faculty of Information Science&Technology Universiti Kebangsaan MalaysiaBangi43600Malaysia Department of Computer Science Applied CollegeUniversity of HailHail55424Saudi Arabia Department of Computer Science and Informatics Taibah UniversityMedina42353Saudi Arabia Department of Computer Science Hanyang UniversitySeoul04763Republic of Korea

This research aims to enhance Clinical Decision Support Systems(CDSS)within Wireless Body Area Networks(WBANs)by leveraging advanced machine learning ***,we target the challenges of accurate diagnosis in medical imaging and sequential data analysis using Recurrent Neural Networks(RNNs)with Long Short-Term Memory(LSTM)layers and echo state *** models are tailored to improve diagnostic precision,particularly for conditions like rotator cuff tears in osteoporosis patients and gastrointestinal *** diagnostic methods and existing CDSS frameworks often fall short in managing complex,sequential medical data,struggling with long-term dependencies and data imbalances,resulting in suboptimal accuracy and delayed *** goal is to develop Artificial Intelligence(AI)models that address these shortcomings,offering robust,real-time diagnostic *** propose a hybrid RNN model that integrates SimpleRNN,LSTM layers,and echo state cells to manage long-term dependencies ***,we introduce CG-Net,a novel Convolutional Neural Network(CNN)framework for gastrointestinal disease classification,which outperforms traditional CNN *** further enhance model performance through data augmentation and transfer learning,improving generalization and robustness against data scarcity and *** validation,including 5-fold cross-validation and metrics such as accuracy,precision,recall,F1-score,and Area Under the Curve(AUC),confirms the models’***,SHapley Additive exPlanations(SHAP)and Local Interpretable Model-agnostic Explanations(LIME)are employed to improve model *** findings show that the proposed models significantly enhance diagnostic accuracy and efficiency,offering substantial advancements in WBANs and CDSS.

关键词： computer science clinical decision support system(CDSS) medical queries healthcare deep learning recurrent neural network(RNN) long short-term memory(LSTM)

来源：评论

学校读者我要写书评

暂无评论

Curious Administrator's Misbehaviors in Message Layer Security (MLS)

Curious Administrator's Misbehaviors in Message Layer Securi...

引用

2025 International Conference on Computing, Networking and Communications, ICNC 2025

作者： Choi, Hyoung-Kee Kwon, Songhui Sungkyunkwan University College of Computing and Informatics Suwon Korea Republic of Sungkyunkwan University Department of Electrical and Computer Engineering Suwon Korea Republic of

ISBN: (纸本)9798331520960

A key factor of the MLS group protocol is that it provides asynchronously efficient group key establishment with forward secrecy and post-compromise security for many groups in a standardized way. End-to-end encryption (E2EE) is essential to fulfilling the correctness of security in a messenger application. The correctness heavily depends on the trustworthiness of a middleman server in the center where a curious administrator may break E2EE. If this happened the administrator as a nonmember could read messages and impersonate a member by intervening with group-key management. This paper presents an in-depth explanation of the MLS protocol and demonstrates a proof-of-concept of curious administrator's misbehaviors by analyzing the protocol. Our research outputs suggest that unconditionally trustworthy relationships on the central server in designing network protocols for security needs to prepare for any incidents caused by misbehaving administrator of the central server. © 2025 IEEE.

关键词： End-to-End Encryption Message Layer Security Secure Messaging Service

来源：评论

学校读者我要写书评

暂无评论

CEGSet: Collection of standardized cause-effect graph specifications 12

CEGSet: Collection of standardized cause-effect graph specif...

引用

12th Mediterranean Conference on Embedded Computing, MECO 2023

作者： Krupalija, Ehlimana Cogo, Emir Becirovic, Seila Prazina, Irfan Pozderac, Damir Besic, Ingmar University of Sarajevo Faculty of Electrical Engineering Department of Computer Science and Informatics Sarajevo Bosnia and Herzegovina

ISBN: (纸本)9798350322910

Cause-effect graphs are a commonly used black-box testing method, and many different algorithms for converting system requirements to cause-effect graph specifications and deriving test case suites have been proposed. However, in order to test the efficiency of black-box testing algorithms on a variety of cause-effect graphs containing different numbers of nodes, logical relations and dependency constraints, a dataset containing a collection of cause-effect graph specifications created by authors of existing papers is necessary. This paper presents CEGSet, the first collection of existing cause-effect graph specifications. The dataset contains a total of 65 graphs collected from the available relevant literature. The specifications were created by using the ETF-RI-CEG graphical software tool and can be used by future authors of papers focusing on the cause-effect graphing technique. The collected graphs can be re-imported in the tool and used for the desired purposes. The collection also includes the specification of system requirements in the form of natural language from which the cause-effect graphs were derived where possible. This will encourage future work on automatizing the process of converting system requirements to cause-effect graph specifications. © 2023 IEEE.

关键词： Black-box testing

来源：评论

学校读者我要写书评

暂无评论

Optimization of the test case minimization algorithm based on forward-propagation in cause-effect graphs 22

Optimization of the test case minimization algorithm based o...

引用

22nd International Symposium INFOTEH-JAHORINA, INFOTEH 2023

作者： Krupalija, Ehlimana Cogo, Emir Bećirović, Šeila Prazina, Irfan Pozderac, Damir Bešić, Ingmar University of Sarajevo Faculty of Electrical Engineering Department of Computer Science and Informatics Sarajevo Bosnia and Herzegovina

ISBN: (纸本)9781665475464

Many different methods are used for generating blackbox test case suites. Test case minimization is used for reducing the feasible test case suite size in order to minimize the cost of testing while ensuring maximum fault detection. This paper presents an optimization of the existing test case minimization algorithm based on forward-propagation of the cause-effect graphing method. The algorithm performs test case prioritization based on test case strength, a newly introduced test case selection metric. The optimized version of the minimization algorithm was evaluated by using thirteen different examples from the available literature. In cases where the existing algorithm did not generate the minimum test case subsets, significant improvements of test effect coverage metric values were achieved. Test effect coverage metric values were not improved only in cases where maximum optimization was already achieved by using the existing algorithm. © 2023 IEEE.

关键词： Black-box testing

来源：评论

学校读者我要写书评

暂无评论

Open-Set Heterogeneous Domain Adaptation: Theoretical Analysis and Algorithm 39

Open-Set Heterogeneous Domain Adaptation: Theoretical Analys...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Pham, Thai-Hoang Wang, Yuanlong Yin, Changchang Zhang, Xueru Zhang, Ping Department of Computer Science and Engineering The Ohio State University United States Department of Biomedical Informatics The Ohio State University United States

ISBN: (纸本)157735897X

Domain adaptation (DA) tackles the issue of distribution shift by learning a model from a source domain that generalizes to a target domain. However, most existing DA methods are designed for scenarios where the source and target domain data lie within the same feature space, which limits their applicability in real-world situations. Recently, heterogeneous DA (HeDA) methods have been introduced to address the challenges posed by heterogeneous feature space between source and target domains. Despite their successes, current HeDA techniques fall short when there is a mismatch in both feature and label spaces. To address this, this paper explores a new DA scenario called open-set HeDA (OSHeDA). In OSHeDA, the model must not only handle heterogeneity in feature space but also identify samples belonging to novel classes. To tackle this challenge, we first develop a novel theoretical framework that constructs learning bounds for prediction error on target domain. Guided by this framework, we propose a new DA method called Representation Learning for OSHeDA (RL-OSHeDA). This method is designed to simultaneously transfer knowledge between heterogeneous data sources and identify novel classes. Experiments across text, image, and clinical data demonstrate the effectiveness of our algorithm. Copyright © 2025, Association for the Advancement of Artificia Intelligence (***). All rights reserved.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

A Low-Cost Real-Time Cyber Physical System for Overcoming Excess Braking Issues In Race Cars 12

A Low-Cost Real-Time Cyber Physical System for Overcoming Ex...

引用

12th International Conference on Modern Circuits and Systems Technologies, MOCAST 2023

作者： Samaras, V. Tatas, K. Lontos, A. Frederick University Department of Electrical Engineering Computer Engineering and Informatics Nicosia Cyprus Frederick University Department of Mechanical Engineering Nicosia Cyprus

ISBN: (纸本)9798350321074

This paper presents the development of proof of concept for a low-cost, real-time Cyber Physical System (CPS)/IoT that deals with observed excess braking issues in Formula Student (FS) race cars. The goal is to utilize telemetry aspects through a smart monitoring system under the FS low budget competition restrictions, with low-cost materials, open-source software, Long Range wireless communication (LoRa) and cloud services. The paper also introduces an algorithm for implementing an adaptive sampling rate based on the proximity of monitored parameters to warning levels to minimize unnecessary data transmission. The system prototype was tested under realistic driving conditions, obtaining useful preliminary data that can be used to guide both driving behavior and future vehicle design. The results demonstrate that the proposed approach can be useful for the race team in FS conditions and may be possible to expand it to other similar driving scenarios outside FS. © 2023 IEEE.

关键词： Cyber Physical System

来源：评论

学校读者我要写书评

暂无评论

Developing a Green AI Platform for SMEs to Analyze Open-source Information with the Approach of Society 5.0

引用

SN computer Science 2024年第7期5卷 923页

作者： Akarslan, Hüseyin Sağiroğlu, Şeref Management Information Systems Graduate School of Informatics Gazi University Türkiye Ankara Turkey Computer Engineering Department Engineering Faculty Gazi University Türkiye Ankara Turkey

Management and information systems are considered two major disciplines that have to be combined to benefit from knowledge for rational decision-making which means strict procedures utilizing objective knowledge and logic for many years. Effective and efficient use of already invested organizational resources is a key subject to gain a competitive advantage by making decisions with the help of AI techniques which benefit from both internal and external knowledge. This paper introduces a new model to get benefits from open-source intelligence (OSINT) in the context of external knowledge with a green computing approach for solving global organizational challenges in the scope of Society 5.0. This research has a dual background, theoretical and practical. Problem defining and hypothesis formulation was carried out by literature review. Then, the application developed according to the proposed model was put into action in some organizations for testing, and feedback from experts who participated in the proof of concept (POC) activities was evaluated and discussed empirically. Finally, the findings of our research, the advantages of our model, and both academic and business future work were summarized. It has been practically determined that the model and application we have developed are largely successful in line with our hypotheses. However, the weaknesses and advantages of our study were also revealed. Our research is the first study that combines Management Information Systems and Open-Source Intelligence disciplines with the approach of Society 5.0 to support SMEs in analyzing open-source information to build better decision-making mechanisms. The developed application in this study can also be applied to solving other green distributed computing needs in different domains. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Distributed computing External knowledge Green computing Management information systems (MIS) Open-source intelligence (OSINT) Society 5.0

来源：评论

学校读者我要写书评

暂无评论

Evaluating Beam Sweeping for AoA Estimation with an RIS Prototype: Indoor/Outdoor Field Trials

Evaluating Beam Sweeping for AoA Estimation with an RIS Prot...

引用

2025 IEEE Wireless Communications and Networking Conference, WCNC 2025

作者： Vordonis, Dimitris Kompostiotis, Dimitris Paliouras, Vassilis Alexandropoulos, George C. Grec, Florin University of Patras Electrical and Computer Engineering Department Greece National and Kapodistrian University of Athens Department of Informatics and Telecommunications Greece Noordwijk Netherlands

ISBN: (纸本)9798350368369

Reconfigurable Intelligent Surfaces (RISs) have emerged as a promising technology to enhance wireless communication systems by enabling dynamic control over the propagation environment. However, practical experiments are crucial towards the validation of the theoretical potential of RISs while establishing their real-world applicability, especially since most studies rely on simplified models and lack comprehensive field trials. In this paper, we present an efficient method for configuring a 1-bit RIS prototype at sub-6 GHz, resulting in a codebook oriented for beam sweeping;an essential protocol for initial access and Angle of Arrival (AoA) estimation. The measured radiation patterns of the RIS validate the theoretical model, demonstrating consistency between the experimental results and the predicted beamforming behavior. Furthermore, we experimentally prove that RIS can alter channel properties and by harnessing the diversity it provides, we evaluate beam sweeping as an AoA estimation technique. Finally, we investigate the frequency selectivity of the RIS and propose an approach to address indoor challenges by leveraging the geometry of environment. © 2025 IEEE.

关键词： beam sweeping field trials MIMO prototype radiation pattern Reconfigurable intelligent surfaces

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：