检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2024年第9期5卷 4322-4343页

作者： Al-Maliki, Shawqi Qayyum, Adnan Ali, Hassan Abdallah, Mohamed Qadir, Junaid Hoang, Dinh Thai Niyato, Dusit Al-Fuqaha, Ala Division College of Science and Engineering Doha34110 Qatar Information Technology University Lahore54000 Pakistan University of New South Wales Sydney Sydney2052 Australia Qatar University Department of Computer Science and Engineering College of Engineering Doha34110 Qatar University of Technology Sydney School of Electrical and Data Engineering Sydney2052 Australia Nanyang Technological University College of Computing and Data Science 639798 Singapore

Deep neural networks (DNNs) have been the driving force behind many of the recent advances in machine learning. However, research has shown that DNNs are vulnerable to adversarial examples - input samples that have been perturbed to force DNN-based models to make errors. As a result, adversarial machine learning (AdvML) has gained a lot of attention, and researchers have investigated these vulnerabilities in various settings and modalities. In addition, DNNs have also been found to incorporate embedded bias and often produce unexplainable predictions, which can result in antisocial AI applications. The emergence of new AI technologies that leverage large language models (LLMs), such as ChatGPT and GPT-4, increases the risk of producing antisocial applications at scale. AdvML for social good (AdvML4G) is an emerging field that repurposes the AdvML bug to invent prosocial applications. Regulators, practitioners, and researchers should collaborate to encourage the development of prosocial applications and hinder the development of antisocial ones. In this work, we provide the first comprehensive review of the emerging field of AdvML4G. This paper encompasses a taxonomy that highlights the emergence of AdvML4G, a discussion of the differences and similarities between AdvML4G and AdvML, a taxonomy covering social good-related concepts and aspects, an exploration of the motivations behind the emergence of AdvML4G at the intersection of ML4G and AdvML, and an extensive summary of the works that utilize AdvML4G as an auxiliary tool for innovating prosocial applications. Finally, we elaborate upon various challenges and open research issues that require significant attention from the research community. © 2020 IEEE.

关键词： Taxonomies

来源：评论

学校读者我要写书评

暂无评论

MLSTM: A Novel Modified Long Short-Term Memory Networks for Melody Synthesis Melody Generation

MLSTM: A Novel Modified Long Short-Term Memory Networks for ...

引用

International Conference on Advanced Communications and Machine Intelligence, MICA 2023

作者： Ganguly, Tanudeep Sharma, Sushmita Deb, Suman Ezugwu, Absalom El-Shamir Saha, Apu Kumar Department of Computer Science and Engineering National Institute of Technology Agartala Tripura Agartala India Department of Computer Science and Engineering Indian Institute of Information Technology Agartala Tripura Agartala India Unit for Data Science and Computing North-West University Potchefstroom2520 South Africa

ISBN: (纸本)9789819762217

This research elaborates a melody synthesis model utilizing a Modified Long Short-Term Memory (mLSTM) neural network, trained on Kern dataset, preprocessed into Musical Instrument Digital Interface (MIDI) files. After receiving a seed melody as input 01, the model predicts the note that will come next by looking at the notes that have come before it. The model can learn the structures and patterns seen in classical music by using MIDI files as the training dataset. Categorical cross-entropy loss function is used to train the model over several epochs with early halting. To regulate how random the predictions are, the model makes use of a temperature parameter. The assessment demonstrates that the produced tunes have a variety of traits and are aesthetically pleasing to human ears. To justify the performance of the proposed method some criteria were considered and expert evaluation was done. The evaluation shows that the model has generated the melodies that are pleasing to the ear and have a similar structure to the training dataset. The model has several uses, including prediction and music composition. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Sublinear-Space Streaming Algorithms for Estimating Graph Parameters on Sparse Graphs 1

引用

18th International Symposium on Algorithms and data Structures, WADS 2023

作者： Chen, Xiuge Chitnis, Rajesh Eades, Patrick Wirth, Anthony School of Computing and Information Systems The University of Melbourne ParkvilleVIC Australia School of Computer Science University of Birmingham Birmingham United Kingdom

ISBN: (数字)9783031389061

ISBN: (纸本)9783031389054

In this paper, we design sub-linear space streaming algorithms for estimating three fundamental parameters – maximum independent set, minimum dominating set and maximum matching – on sparse graph classes, i.e., graphs which satisfy m= O(n) where m, n is the number of edges, vertices respectively. Each graph parameter we consider can have size Ω(n) even on sparse graph classes, and hence for sublinear-space algorithms we are restricted to parameter estimation instead of attempting to find a solution. We obtain these results: Estimating Max Independent Set via the Caro-Wei bound: Caro and Wei each showed λ= ∑ v1 / (d(v) + 1 ) is a lower bound on max independent set size, where vertex v has degree d(v). If average degree, d¯, is O(1 ), and max degree Δ= O(Ε2d¯ - 3n), our algorithms, with at least 1 - δ success probability:In online streaming, return an actual independent set of size 1 ± Ε times λ. This improves on Halldórsson et al. [Algorithmica ’16]: we have less working space, i.e., O(log Ε- 1· log n· log δ- 1), faster updates, i.e., O(log Ε- 1), and bounded success *** insertion-only streams, approximate λ within factor 1 ± Ε, in one pass, in O(d¯ Ε- 2log n· log δ- 1) space. This aligns with the result of Cormode et al. [ISCO ’18], though our method also works for online streaming. In a vertex-arrival and random-order stream, space reduces to O(log (d¯ Ε- 1) ). With extra space and post-processing step, we remove the max-degree ***-Space Algorithms on Forests: On a forest, Esfandiari et al. [SODA ’15, TALG ’18] showed space lower bounds for 1-pass randomized algorithms that approximately estimate these graph parameters. We narrow the gap between upper and lower bounds:Max independent set size within 3 / 2 · (1 ± Ε) in one pass and in log O(1)n space, and within 4 / 3 · (1 ± Ε) in two passes and in O~(n) space;the lower bound is for approx. ≤ 4 / *** dominating set size within 3 · (1 ± Ε) in one pass and in log O(1)n space, and within 2

关键词： Parameter estimation

来源：评论

学校读者我要写书评

暂无评论

Lightweight Authenticated Key Agreement Protocol for Smart Power Grid systems Using PUF

IEEE Open Journal of the Communications Society

引用

IEEE Open Journal of the Communications Society 2024年 5卷 3568-3580页

作者： Zahoor, Amina Mahmood, Khalid Saleem, Muhammad Asad Badar, Hafiz Muhammad Sanaullah Le, Tuan-Vinh Das, Ashok Kumar Department of Computer Science Sahiwal57000 Pakistan National Yunlin University of Science and Technology Graduate School of Intelligent Data Science Douliu64002 Taiwan University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu611731 China Emerson University Multan Department of Emerging Computing Technologies Multan61000 Pakistan Fu Jen Catholic University Bachelor's Program of Artificial Intelligence and Information Security New Taipei City24206 Taiwan International Institute of Information Technology Hyderabad Center for Security Theory and Algorithmic Research Hyderabad500032 India

The Smart Power Grid (SPG) is pivotal in orchestrating and managing demand response in contemporary smart cities, leveraging the prowess of information and Communication Technologies (ICTs). Within the immersive SPG environment, the ubiquitous deployment of smart meters stands as a testament to their paramount importance in the realm of vigilant monitoring and oversight. These smart meters are installed on high-tension electricity lines and transmit information about electricity outages and other issues to the utility center. To access services from utility centers, smart meters need to communicate securely over a public channel, even though the network itself is insecure. However, potential attacks from adversaries ( Ad) can exploit this communication. Therefore, protecting this communication is of utmost importance. Several privacy-preserving authentication protocols designed for SPG have been introduced in the literature. Nevertheless, a significant number of these protocols exhibit vulnerabilities to diverse security attacks. This article introduces a lightweight and anonymous authentication protocol specifically designed to address these concerns in the SPG environment. Our protocol ensures both security and efficiency, surpassing other comparable protocols in terms of its lightweight nature. By employing both formal and informal analysis, we showcase the robustness of our protocol against significant attacks while remaining lightweight. The proposed protocol provides added security features and incurs 23.8879% lower computation cost than related protocols. Consequently, our protocol is highly suitable for implementation in the SPG system. © 2020 IEEE.

关键词： Hash functions

来源：评论

学校读者我要写书评

暂无评论

Plausible Heterogeneous Graph k-Anonymization for Social Networks

引用

Tsinghua science and Technology 2022年第6期27卷 912-924页

作者： Kaiyang Li Ling Tian Xu Zheng Bei Hui School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 611731China School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu 610054China Trusted Cloud Computing and Big Data Key Laboratory of Sichuan Province Chengdu 610000China

The inefficient utilization of ubiquitous graph data with combinatorial structures necessitates graph embedding methods,aiming at learning a continuous vector space for the graph which is amenable to be adopted in traditional machine learning algorithms in favor of vector *** embedding methods build an important bridge between social network analysis and data analytics as social networks naturally generate an unprecedented volume of graph data *** social network data not only bring benefit for public health,disaster response,commercial promotion,and many other applications,but also give birth to threats that jeopardize each individual’s privacy and ***,most existing works in publishing social graph embedding data only focus on preserving social graph structure with less attention paid to the privacy issues inherited from social *** be specific,attackers can infer the presence of a sensitive relationship between two individuals by training a predictive model with the exposed social network *** this paper,we propose a novel link-privacy preserved graph embedding framework using adversarial learning,which can reduce adversary’s prediction accuracy on sensitive links while persevering sufficient non-sensitive information such as graph topology and node attributes in graph *** experiments are conducted to evaluate the proposed framework using ground truth social network datasets.

关键词： social network graph embedding privacy preservation adversarial learning

来源：评论

学校读者我要写书评

暂无评论

Fuzz-classification(p,l)-Angel:An enhanced hybrid artificial intelligence based fuzzy logic for multiple sensitive attributes against privacy breaches

引用

Digital Communications and Networks 2023年第5期9卷 1131-1140页

作者： Tehsin Kanwal Hasina Attaullah Adeel Anjum Abid Khan Gwanggil Jeon Department of Computer Science COMSATS University IslamabadPakistan Department of Information Technology Quaid-e-Azam University IslamabadPakistan College of Science and Engineering School of Computing and MathsUniversity of DerbyDE221GBUK Department of Embedded Systems Engineering Incheon National UniversitySouth Korea

The inability of traditional privacy-preserving models to protect multiple datasets based on sensitive attributes has prompted researchers to propose models such as SLOMS,SLAMSA,(p,k)-Angelization,and(p,l)-Angelization,but these were found to be insufficient in terms of robust privacy and performance.(p,l)-Angelization was successful against different privacy disclosures,but it was not *** the best of our knowledge,no robust privacy model based on fuzzy logic has been proposed to protect the privacy of sensitive attributes with multiple *** this paper,we suggest an improved version of(p,l)-Angelization based on a hybrid AI approach and privacy-preserving approach like ***-classification(p,l)-Angel uses artificial intelligence based fuzzy logic for classification,a high-dimensional segmentation technique for segmenting quasi-identifiers and multiple sensitive *** demonstrate the feasibility of the proposed solution by modelling and analyzing privacy violations using High-Level Petri *** results of the experiment demonstrate that the proposed approach produces better results in terms of efficiency and utility.

关键词： Generalization Fuzzy-logic MSA Privacy disclosures Membership function (p,l)-Angelization QT HLPN

来源：评论

学校读者我要写书评

暂无评论

Evaluating Cryptocurrency Market Risk on the Blockchain: An Empirical Study Using the ARMA-GARCH-VaR Model

IEEE Open Journal of the Computer Society

引用

IEEE Open Journal of the computer Society 2024年 5卷 83-94页

作者： Huang, Yongrong Wang, Huiqing Chen, Zhide Feng, Chen Zhu, Kexin Yang, Xu Yang, Wencheng Fujian Normal University School of Computer and Cyberspace Security Fuzhou350117 China Department of Information Engineering Fuzhou Polytechnic Fuzhou350108 China Department of Computer Science and Engineering National Sun Yat-sen University Kaohsiung80424 Taiwan College of Computer and Data Science Minjiang University Fuzhou350108 China School of Mathematics Physics and Computing University of Southern Queensland ToowoombaQLD4350 Australia

Cryptocurrency, a novel digital asset within the blockchain technology ecosystem, has recently garnered significant attention in the investment world. Despite its growing popularity, the inherent volatility and instability of cryptocurrency investments necessitate a thorough risk evaluation. This study utilizes the Autoregressive Moving Average (ARMA) model combined with the Generalized Autoregressive Conditionally Heteroscedastic (GARCH) model to analyze the volatility of three major cryptocurrencies-Bitcoin (BTC), Ethereum (ETH), and Binance Coin (BNB)-over a period from January 1, 2017, to October 29, 2022. The dataset comprises daily closing prices, offering a comprehensive view of the market's fluctuations. Our analysis revealed that the value-at-risk (VaR) curves for these cryptocurrencies demonstrate significant volatility, encompassing a broad spectrum of returns. The overall risk profile is relatively high, with ETH exhibiting the highest risk, followed by BTC and BNB. The ARMA-GARCH-VaR model has proven effective in quantifying and assessing the market risks associated with cryptocurrencies, providing valuable insights for investors and policymakers in navigating the complex landscape of digital assets. © 2020 IEEE.

关键词： Investments

来源：评论

学校读者我要写书评

暂无评论

MeHyper: Accelerating Hypergraph Neural Networks by Exploring Implicit dataflows 31

MeHyper: Accelerating Hypergraph Neural Networks by Explorin...

引用

31st IEEE International Symposium on High Performance computer Architecture, HPCA 2025

作者： Zhao, Wenju Yao, Pengcheng Chen, Dan Zheng, Long Liao, Xiaofei Wang, Qinggang Ma, Shaobo Li, Yu Liu, Haifeng Xiao, Wenjing Sun, Yufei Zhu, Bing Jin, Hai Xue, Jingling Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Wuhan430074 China National University of Singapore School of Computing 119077 Singapore Guangxi University School of Computer Electronics and Information NanNing530004 China University of New South Wales School of Computer Science and Engineering SydneyNSW2052 Australia

ISBN: (纸本)9798331506476

Hypergraph Neural Networks (HGNNs) are increasingly utilized to analyze complex inter-entity relationships. Traditional HGNN systems, based on a hyperedge-centric dataflow model, independently process aggregation tasks for hyperedges and vertices, leading to significant computational redundancy. This redundancy arises from recalculating shared information across different tasks. For the first time, we identify and harness implicit dataflows (i.e., dependencies) within HGNNs, introducing the microedge concept to effectively capture and reuse intricate shared information among aggregation tasks, thereby minimizing redundant computations. We have developed a new microedge-centric dataflow model that processes shared information as fine-grained microedge aggregation tasks. This dataflow model is supported by the Read-Process-Activate-Generate execution model, which aims to optimize parallelism among these tasks. Furthermore, our newly developed MeHyper, a microedge-centric HGNN accelerator, incorporates a decoupled pipeline for improved computational parallelism and a hierarchical feature management strategy to reduce off-chip memory accesses for large volumes of intermediate feature vectors generated. Our evaluation demonstrates that MeHyper substantially outperforms the leading CPUbased system PyG-CPU and the GPU-based system HyperGef, delivering performance improvements of 1,032.23 × and 10.51 ×, and energy efficiencies of 1,169.03 × and 9.96 ×, respectively. © 2025 IEEE.

关键词： Redundancy

来源：评论

学校读者我要写书评

暂无评论

Who Responded to Whom: The Joint Effects of Latent Topics and Discourse in Conversation Structure 10

Who Responded to Whom: The Joint Effects of Latent Topics an...

引用

10th SIGHAN Workshop on Chinese Language Processing, SIGHAN 2024

作者： Ji, Lu Chen, Lei Li, Jing Wei, Zhongyu Zhang, Qi Huang, Xuanjing School of Computer Science Fudan University China School of Data Science Fudan University China Department of Computing The Hong Kong Polytechnic University Hong Kong Research Institute of Intelligent and Complex Systems Fudan University China

ISBN: (纸本)9798891761551

Vast amount of online conversations are produced on a daily basis, resulting in a pressing need to automatic conversation understanding. As a basis to structure a discussion, we identify the responding relations in the conversation discourse, which link response utterances to their initiations. To figure out who responded to whom, here we explore how the consistency of topic contents and dependency of discourse roles indicate such interactions, whereas most prior work ignore the effects of latent factors underlying word occurrences. We propose a neural model to learn latent topics and discourse in word distributions, and predict pairwise initiation-response links via exploiting topic consistency and discourse dependency. Experimental results on both English and Chinese conversations show that our model significantly outperforms the previous state of the arts. © 2024 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

Cloud Intrusion Detection System Based on Honeynet, Honeywell, Honeypot, and Honeytoken Strategies

Cloud Intrusion Detection System Based on Honeynet, Honeywel...

引用

International Conference on data science and Network Engineering, ICDSNE 2023

作者： Yasotha, B. Arthy, M. Shoba, L.K. Loganathan, Muralidaran Department of Data Science and Business Systems SRM Institute of Science and Technology Kattankulathur Tamil Nadu India Department of Computing Technologies SRM Institute of Science and Technology Kattankulathur Tamil Nadu India Computer Science and Information Systems Engineering St. Joseph University Dares Salaam Tanzania United Republic of

ISBN: (纸本)9789819967544

The security aspect of cloud computing is much more challenging for researchers. Preventing the attack requires knowledge about the type of attack, its origin, and how vulnerabilities and tools are used for the attack. The cloud security methodology protects customer data, information, and applications from attackers. Due to digitalization, the volume of data is increased. The protection of data is very challenging for cloud service providers. This work proposes a new cloud intrusion detection system security infrastructure based on Honeynet, Honeywell, Honeypot, and Honeytoken Strategies. The proposed strategy effectively identifies intrusion detection, attack behavior, and attack scenario. The testing of data was carried out in OpenStack environments. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Intrusion detection

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：