检索结果-内蒙古大学图书馆

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Siqi Shen Mengwei Qiu Jun Liu Weiquan Liu Yongquan Fu Xinwang Liu Cheng Wang Fujian Key Lab of Sensing and Computing for Smart Cities School of Informatics Xiamen University China School of Computer National University of Defense Technology China

ISBN: (纸本)9781713871088

The factorization of state-action value functions for Multi-Agent Reinforcement Learning (MARL) is important. Existing studies are limited by their representation capability, sample efficiency, and approximation error. To address these challenges, we propose, ResQ, a MARL value function factorization method, which can find the optimal joint policy for any state-action value function through residual functions. ResQ masks some state-action value pairs from a joint state-action value function, which is transformed as the sum of a main function and a residual function. ResQ can be used with mean-value and stochastic-value RL. We theoretically show that ResQ can satisfy both the individual global max (IGM) and the distributional IGM principle without representation limitations. Through experiments on matrix games, the predator-prey, and StarCraft benchmarks, we show that ResQ can obtain better results than multiple expected/stochastic value factorization methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Analysis the Effect of Clustering and Lightweight Encryption Approaches on WSNs Lifetime 8

Analysis the Effect of Clustering and Lightweight Encryption...

引用

8th International Conference on Electrical Engineering and informatics, ICEEI 2021

作者： Ahmad, Rami Sundararajan, Elankovan A Abu-Ain, Tarik School of Information Technology Sebha University Sebha Libya Universiti Kebangsaan Malaysia Center for Software Technology and Management Faculty of Information Science Technology Selangor Malaysia College of Computing and Informatics Saudi Electronic University Saudi Arabia

ISBN: (纸本)9781665405560

The energy consumption and security efficiency are still main challenges in Wireless Sensor Networks (WSNs) due to their hardware resource-constrained nature. The 6LoWPAN protocol was developed to improve WSNs communication, security, and node management optimization. Hence the protocol energy efficiency and security can be improved. In this paper, we address the WSN nodes' power consumption by analyzing the Dynamic Cluster Head (DynCH) technique, which automates the process of selecting WSN Cluster Head (CH) nodes based on the WSN nodes' energy and nodes' distances among each other in mobile WSN nodes. Moreover, this analysis covers the complexity of DynCH in different environments to prove its efficiency compared to the steady CH mechanism. In addition, we present the performance analysis of different lightweight systematic block encryption algorithms along with DynCH scheme on WSNs lifetime. In particular, Speck128, FlexenTech, Tiny Encryption Algorithm (TEA), and Advanced Encryption Standard (AES) algorithms are used in order to determine the amount of energy consumed by the sensor nodes and their effect on the network lifetime. The Cooja simulator is used to evaluate our analysis. Finally, the outcome of the analysis has showed that DynCH improves the wireless network lifetime by 45% compared to steady clustering approach. Moreover, the analysis also shows that, Speck128 consumed 26%, FlexenTech consumed 52%, TEA consumed 65%, and AES consumed 78% of wireless network lifetime compared to unsecure wireless networks communication, respectively. © 2021 IEEE.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

TAD-Bench: A Comprehensive Benchmark for Embedding-Based Text Anomaly Detection

arXiv

引用

arXiv 2025年

作者： Cao, Yang Yang, Sikun Li, Chen Xiang, Haolong Qi, Lianyong Liu, Bo Li, Rongsheng Liu, Ming School of Computing and Information Technology Great Bay University China Great Bay Institute for Advanced Study Great Bay University China Graduate School of Informatics Nagoya University Japan School of Software Nanjing University of Information Science and Technology China College of Computer Science and Technology China University of Petroleum [East China China College of Cyberspace Security Zhengzhou University China School of Computer Harbin Engineering University China School of IT Deakin University Australia

Text anomaly detection is crucial for identifying spam, misinformation, and offensive language in natural language processing tasks. Despite the growing adoption of embedding-based methods, their effectiveness and generalizability across diverse application scenarios remain under-explored. To address this, we present TAD-Bench, a comprehensive benchmark designed to systematically evaluate embedding-based approaches for text anomaly detection. TAD-Bench integrates multiple datasets spanning different domains, combining state-of-the-art embeddings from large language models with a variety of anomaly detection algorithms. Through extensive experiments, we analyze the interplay between embeddings and detection methods, uncovering their strengths, weaknesses, and applicability to different tasks. These findings offer new perspectives on building more robust, efficient, and generalizable anomaly detection systems for real-world applications. Copyright © 2025, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

An urban-scale spatiotemporal optimization of rooftop photovoltaic charging of electric vehicles

引用

Urban informatics 2024年第1期3卷 1-18页

作者： Ji, Nanfan Zhu, Rui Huang, Ziyi You, Linlin Department of Land Surveying and Geo-Informatics The Hong Kong Polytechnic University Kowloon Hong Kong China Institute of High Performance Computing (IHPC) Agency for Science Technology and Research (A*STAR) Singapore Republic of Singapore School of Intelligent Systems Engineering Sun Yat-Sen University Shenzhen China

Solar photovoltaic (PV) farming is increasingly being used to power electric vehicles (EVs). Although many studies have developed dynamic EV charging prediction and scheduling models, few of them have coupled rooftop PV electricity generation with the spatiotemporal EV charging demands at an urban scale. Thus, this study develops a research framework containing three interconnected modules to investigate the feasibility of EV charging powered by rooftop PVs. The framework is constructed by the statistics of time serial EV charging demands at each station, the planning of rooftop PV installations associated with all charging stations, and the development of a dynamic dispatching algorithm to transmit surplus electricity from one station to another. The algorithm can maximize the overall balance between supply and demand, maximize the total PV electricity generation while minimising the total PV area, minimize the number of PV charging stations used as the suppliers for dynamic dispatch, and minimize the total electricity transmission distance between stations given the same power supply. The experiment utilizes a complete EV charging dataset containing 5574 charging piles with more than 9.7 million records in June and July in Guangzhou, China. The results show that rooftop PVs can supply more than 90% of the charging demand. The results encourage and inspire us to generalize and promote such a solution in other cities. Future work can refine the algorithm by adapting different PV sizes into various charging stations to further improve the electricity generation capability and the dynamic dispatching efficiency.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Blockchain Trust Mechanism in Named Data Networking: A Systematic Literature Review 6

Blockchain Trust Mechanism in Named Data Networking: A Syste...

引用

6th International Conference on Internet Applications, Protocols and Services, NETAPPS 2020

作者： Rosli, Athirah Hassan, Suhaidi Omar, Mohd Hasbullah Faculty of Ocean Engineering Technology and Informatics Universiti Malaysia Terengganu Kuala Terengganu Malaysia InterNetWorks Research Laboratory School of Computing Universiti Utara Malaysia Sintok Kedah Malaysia

Blockchain has been reported as one of the technologies that could change the Internet architecture in the future. By offering better mechanism that increases trust in the network, it gives possibilities for the future Internet, Named Data Networking (NDN) to adapt such mechanism. This paper focuses to look into the potential trust mechanisms of Blockchain to be adapted in NDN. Systematic Literature Review (SLR) is used as a technique to dig into current researches to identify research that supports the research question. From the findings, the consensus mechanism has been seen as the potential mechanism that offers trust in Blockchain. © 2021 Proceedings of 6th International Conference on Internet Applications, Protocols and Services: Enduring Internet's Use and Growth in the Pandemic Era, NETAPPS 2020. All Rights Reserved.

关键词： Blockchain consensus mechanism Named Data Networking trust mechanism

来源：评论

学校读者我要写书评

暂无评论

Industry 4.0 (IR 4.0) for Pharmaceutical Industry: An Exploratory Review

Industry 4.0 (IR 4.0) for Pharmaceutical Industry: An Explor...

引用

Visualization, informatics and technology Conference (IVIT), International

作者： Amalia Mukhlas Bazilah A. Talip Jawahir Che Mustapha Shahrinaz Ismail Informatics and Analytics Malaysian Institute of Information Technology Universiti Kuala Lumpur Kuala Lumpur Malaysia School of Computing and Informatics Albukhary International University Alor Setar Kedah Malaysia

ISBN: (纸本)9781665486491

The pharmaceutical industry facing challenges despite digital transformation applying for the benefit of the industry. The companies and the industrial process need to be ready and adapt to the rapid changes or else be left behind by the competitors. The study provides a preliminary understanding of IR 4.0 within the pharmaceutical industry domain. The findings highlight the discussion of IR 4.0 technology in the pharmaceutical industry and the areas in the technology in which the technology has been applied. The research recommends the IR 4.0 technology practices in the operation and process, the elements of technology in the areas, challenges of technology implementation, relatable concepts and theories facilitating the future study.

关键词： Industries Drugs Visualization Systematics Service robots Supply chains Transforms

来源：评论

学校读者我要写书评

暂无评论

A Model Transformation Approach for Converting a Trait-based Model to a Markov Chain Model

A Model Transformation Approach for Converting a Trait-based...

引用

Recent Advances and Innovations in Engineering (ICRAIE)

作者： Nur Diana Madinah Binti Ab Hadi Azlan Ismail Norizan Mat Diah Mastura Md Saad School of IT Faculty of Engineering Built Environment & Information Technology SEGI University & Colleges Petaling Jaya Selangor Malaysia School of Computing Sciences College of Computing Informatics and Mathematics Universiti Teknologi MARA (UiTM) Shah Alam Selangor Malaysia Informatics and Mathematics Institute for Big Data Analytics and Artificial Intelligence (IBDAAI)) University Teknologi MARA (UiTM) Shah Alam Selangor Malaysia

This paper discusses the difficulty of transforming a trait-based model to a Markov Chain Model when the source model is a behavioral model. Therefore, we propose an innovative approach aimed at transforming trait-based models into Markov Chain Models. Our method hinges on the utilization of the Trait Component Ensemble Language (TCOEL) and the establishment of a unified model transformation specification at the meta-model level. Through this approach, we facilitate the seamless conversion of trait-based models into Markov Chain Models, thereby enhancing the adaptability and scope of the original models. The trait-based model presents quite challenging for transforming the behavioral traits directly into verification language. To overcome this difficulty, we analyzed the language behaviors using the fire brigade case study. Then, we used the description to identify and map the appropriate component in order to develop a CTMC model. Finally, we used PRISM model checker to undertake verification.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Circumventing Connectivity for Kernelization 12th

Circumventing Connectivity for Kernelization

引用

12th International Conference on Algorithms and Complexity, CIAC 2021

作者： Jain, Pallavi Kanesh, Lawqueen Roy, Shivesh Kumar Saurabh, Saket Sharma, Roohani Indian Institute of Technology Jodhpur Jodhpur India School of Computing National University of Singapore Singapore Singapore Eindhoven University of Technology Eindhoven Netherlands Institute of Mathematical Sciences HBNI Chennai India Department of Informatics University of Bergen Bergen Norway Max Planck Institute for Informatics Saarland Informatics Campus Saarbrücken Germany

ISBN: (纸本)9783030752415

Classical vertex subset problems demanding connectivity are of the following form: given an input graph G on n vertices and an integer k, find a set S of at most k vertices that satisfies a property and G[S] is connected. In this paper, we initiate a systematic study of such problems under a specific connectivity constraint, from the viewpoint of Kernelization (Parameterized) Complexity. The specific form that we study does not demand that G[S] is connected but rather G[S] has a closed walk containing all the vertices in S. In particular, we study Closed Walk-Subgraph Vertex Cover (CW-SVC, in short), where given a graph G, a set X⊆ E(G), and an integer k;the goal is to find a set of vertices S that hits all the edges in X and can be traversed by a closed walk of length at most k in G. When X is E(G), this corresponds to Closed Walk-Vertex Cover (CW-VC, in short). One can similarly define these variants for Feedback Vertex Set, namely Closed Walk-Subgraph Feedback Vertex Set (CW-SFVS, in short) and Closed Walk-Feedback Vertex Set (CW-FVS, in short). Our results are as follows: CW-VC and CW-SVC both admit a polynomial kernel, in contrast to Connected Vertex Cover that does not admit a polynomial kernel unless NP⊆ coNP/ ***-FVS admits a polynomial kernel. On the other hand CW-SFVS does not admit even a polynomial Turing kernel unless the polynomial-time hierarchy collapses. We complement our kernelization algorithms by designing single-exponential FPT algorithms – 2 O(k)nO(1) – for all the problems mentioned above. © 2021, Springer Nature Switzerland AG.

关键词： Polynomial approximation

来源：评论

学校读者我要写书评

暂无评论

FastAGEDs: Fast Approximate Graph Entity Dependency Discovery 24th

FastAGEDs: Fast Approximate Graph Entity Dependency Discover...

引用

24th International Conference on Web Information Systems Engineering, WISE 2023

作者： Zhou, Guangtong Kwashie, Selasi Zhang, Yidi Bewong, Michael Nofong, Vincent M. Hu, Junwei Cheng, Debo He, Keqing Liu, Shanmei Feng, Zaiwen College of Informatics Huazhong Agricultural University Hubei Wuhan China AI & Cyber Futures Institute Charles Sturt University Bathurst Australia School of Computing Mathematics & Engineering Charles Sturt University Wagga Wagga Australia Department of Computer Science & Engineering University of Mines & Technology Tarkwa Ghana STEM University of South Australia Adelaide Australia School of Computing Science Wuhan University Hubei Wuhan China Hubei Hongshan Laboratory Huazhong Agricultural University Hubei Wuhan China Macro Agricultural Research Institute Huazhong Agricultural University Hubei Wuhan China

ISBN: (纸本)9789819972531

This paper studies the discovery of approximate rules in property graphs. First, we propose a semantically meaningful measure of error for mining graph entity dependencies (GEDs) that almost hold, to tolerate errors and inconsistencies that exist in real-world graphs. Second, we present a new characterisation of GED satisfaction, and devise a depth-first search strategy to traverse the search space of candidate GEDs efficiently. Further, we perform experiments to demonstrate the feasibility and scalability of our solution, FastAGEDs, with three real-world graphs. The results show FastAGEDs is effective and efficient for mining approximate GEDs in noisy and erroneous real-world graphs. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： approximate dependency efficient algorithm graph constraints graph entity dependency

来源：评论

学校读者我要写书评

暂无评论

Mycological Examination of Microscopic Fungi Images with Deep Learning and Gradient Weighted Class Activation Mapping Visualization

Mycological Examination of Microscopic Fungi Images with Dee...

引用

Advances in Science and Engineering technology International Conferences (ASET)

作者： Md. Safinur Rashid Md. Samin Morshed Muhammad Usama Islam Sami Rashid Asif Mahmud Ashraful Islam Department of CSE Islamic University of Technology Dhaka Bangladesh Department of CS University of Arkansas at Little Rock Arkansas USA School of Computing and Informatics University of Louisiana at Lafayette Louisiana USA CCDS Independent University Bangladesh Dhaka Bangladesh

ISBN: (数字)9798350344134

ISBN: (纸本)9798350344141

Fungal species classification is an important task with broad implications for medicine, agriculture, and ecology. Even though they work well, traditional identification techniques frequently take a lot of time, knowledge, and resources. In order to improve the effectiveness and precision of fungal classification, this work explores the use of deep learning techniques, more especially transfer learning with pre-trained Convolutional Neural Networks (CNNs). Using six well-known pre-trained models, namely ResNet-50, ResNet-101, VGG16, VGG19, InceptionV3, and InceptionV4,we modify them to categorize pictures from the DeFungi dataset, which is a varied compilation of images featuring fungal species. Our approach comprises a thorough comparative analysis based on metrics for F1 score, accuracy, and precision to assess how well these models perform in the identification of fungal species. The outcomes show that there is a lot of promise for using these trained models in mycology, with some models exhibiting better performance that might facilitate the creation of automated systems for fungal identification. This work not only advances the field of biological classification by illuminating the strengths and weaknesses of different CNN architectures for fungal classification, but it also emphasizes how crucial it is to use cutting-edge AI methods to tackle biological problems. ResNet-101 was the most accurate with accuracy result of 93 percentile of the six pre-trained models that were examined (ResNet-50, ResNet-101, VGG16, VGG19, InceptionV3, and InceptionV4).Other evaluation metrics including precision was 0.93, recall was 0.90, and F1 score was 0.91. Our results point to a promising direction for future research in automated fungal classification and provide insightful information for researchers and practitioners wishing to use machine learning in the identification and study of fungal species.

关键词： Deep learning Measurement Fungi Visualization Accuracy Biological system modeling Transfer learning Transforms Convolutional neural networks Reliability

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：