检索结果-内蒙古大学图书馆

Proceedings of the 41st International Conference on Machine Learning

作者： Yunke Wang Minjing Dong Yukun Zhao Bo Du Chang Xu School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence and Wuhan Institute of Data Intelligence Wuhan University China Department of Computer Science City University of Hong Kong China School of Computer Science Faculty of Engineering The University of Sydney Australia

Imitation learning has emerged as a promising approach for addressing sequential decision-making problems, with the assumption that expert demonstrations are optimal. However, in real-world scenarios, most demonstrations are often imperfect, leading to challenges in the effectiveness of imitation learning. While existing research has focused on optimizing with imperfect demonstrations, the training typically requires a certain proportion of optimal demonstrations to guarantee performance. To tackle these problems, we propose to purify the potential noises in imperfect demonstrations first, and subsequently conduct imitation learning from these purified demonstrations. Motivated by the success of diffusion model, we introduce a two-step purification via diffusion process. In the first step, we apply a forward diffusion process to smooth potential noises in imperfect demonstrations by introducing additional noise. Subsequently, a reverse generative process is utilized to recover the optimal demonstration from the diffused ones. We provide theoretical evidence supporting our approach, demonstrating that the distance between the purified and optimal demonstration can be bounded. Empirical results on MuJoCo and RoboSuite demonstrate the effectiveness of our method from different aspects.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Probabilistic Framework for Pruning Transformers Via a Finite Admixture of Keys 48

A Probabilistic Framework for Pruning Transformers Via a Fin...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Nguyen, Tan M. Nguyen, Tam Bui, Long Do, Hai Nguyen, Duy Khuong Le, Dung D. Tran-The, Hung Ho, Nhat Osher, Stan J. Baraniuk, Richard G. University of California Department of Mathematics Los Angeles United States Rice University Department of Electrical and Computer Engineering Houston United States Fpt Software Ai Center Ha Noi Viet Nam Vin University College of Engineering and Computer Science Viet Nam Deakin University Applied Artificial Intelligence Institute Geelong Australia The University of Texas Department of Statistics and Data Sciences Austin United States

ISBN: (纸本)9781728163277

Pairwise dot product-based self-attention is key to the success of transformers which achieve state-of-the-art performance across a variety of applications in language and vision, but are costly to compute. It has been shown that most attention scores and keys in transformers are redundant and can be removed without loss of accuracy. In this paper, we develop a novel probabilistic framework for pruning attention scores and keys in transformers. We first formulate an admixture model of attention keys whose input data to be clustered are attention queries. We show that attention scores in self-attention correspond to the posterior distribution of this model when attention keys admit a uniform prior distribution. We then relax this uniform prior constraint and let the model learn these priors from data, resulting in a new Finite Admixture of Keys (FiAK). The learned priors are used for pruning away redundant attention scores and keys in the baseline transformers, improving the diversity of attention patterns that the models capture. We corroborate the efficiency of transformers pruned with FiAK on the ImageNet object classification and WikiText-103 language modeling tasks. Our experiments demonstrate that transformers pruned with FiAK yield similar or better accuracy than the baseline dense transformers while being much more efficient in terms of memory and computational cost. © 2023 IEEE.

关键词： admixture models pruning Transformers

来源：评论

学校读者我要写书评

暂无评论

Socially-Critical software Systems: Is Extended Regulation Required? 1

引用

29th European Systems, software and Services Process Improvement, EuroSPI 2022

作者： Dagg, Niall Kostick, Conor Fallon, James O’Neill, Alex Yilmaz, Murat Messnarz, Richard Clarke, Paul M. School of Computing Dublin City University Dublin Ireland Department of Computer Engineering Gazi University Ankara Turkey ISCN the International Software Consulting Network Graz Austria Lero the Science Foundation Ireland Research Center for Software Limerick Ireland

ISBN: (数字)9783031155598

ISBN: (纸本)9783031155581

Data has become a prevailing aspect of our daily lives, becoming ever more present since the beginning of the 21st century. It is a commodity in today’s world and the amount of data being produced has increased enormously. One of the major ways data is produced and collected is from the use of websites and web-based applications. This data is later used for many different purposes. This paper presents findings from a multivocal literature review, exploring the methods of how this data is collected, what the data is used for once it has been collected, the ethics of data and its collection, and the future of data collection. Among the possible futures, we introduce the concept of socially-critical applications, where data harvesting in web-based applications might require pre-market disclosure and evaluation by notified bodies (instructed by regulation) as a means to break the existing cycle of technology companies outpacing under-resourced and ill-equipped regulators. Rather than regulators continually falling short of enacting laws to satisfy the common good, a new class of socially-critical application could be created in law to permit pre-market evaluation of applications (or versions of applications) that could undermine or interrupt the common good. © 2022, Springer Nature Switzerland AG.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection System Using PSO and DE Algorithms with BP Neural Network 8

Intrusion Detection System Using PSO and DE Algorithms with ...

引用

8th International Conference on Fuzzy Systems and Data Mining, FSDM 2022

作者： Wang, Ke Guan, Jingwei Luo, Song Guan, Zhi MoE Beijing China Department of Computer Science and Technology Eecs Peking University Beijing China Liangjiang International College Chongqing University of Technology Chongqing China School of Computer Science and Engineering Chongqing University of Technology Chongqing China National Engineering Research Center for Software Engineering Peking University Beijing China

ISBN: (纸本)9781643683461

Intrusion detection system (IDS) combines software and hardware to detect network attacks. In this paper, we propose a new intrusion detection method based on an improved BP neural network algorithm. We improve the BP neural network algorithm by combining it with the particle swarm optimization (PSO) algorithm and the differential evolution (DE) algorithm. We also propose a new framework based on common intrusion detection framework to accommodate our improved BP neural network. The experiments based on the CICIDS2017 dataset show our approach achieves better detection efficiency. © 2022 The authors and IOS Press.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

Playing the video games during COVID-19 pandemic and its effects on player's well-being 2

Playing the video games during COVID-19 pandemic and its eff...

引用

2nd International Conference on Business Analytics for Technology and Security, ICBATS 2023

作者： Noaman, Samar Billi Ibrahim, Amer Ali, Liaqat Iqbal, M.W. Ashraf, Asma Haseeb, Usama Muneer, Salman Almajed, Rasah Hamid, K. Canadian University Dubai Faculty of Management Dubai United Arab Emirates American University in the Emirates College of Computer and Information Technology Dubai United Arab Emirates University of Science and Technology of Fujairah College of Engineering and Technology Fujairah United Arab Emirates Superior University Department of Software Engineering Lahore54000 Pakistan Superior University Department of Computer Science Lahore54000 Pakistan NCBA&E School of Computer Science Lahore Pakistan Applied Science Private University Applied Science Research Center Amman11937 Jordan American University in the Dubai College of Computer and Information Technology United Arab Emirates

ISBN: (纸本)9798350335644

COVID-19 affected our lives intensly. That state of affairs made humans helpless. They had been depressed and experienced loneliness. At that time many human beings were determined to play video games just to loosen up their minds. Many games changed into an additional source of revenue wherein during COVID-19 they were playing while earning money. With these advantages, there were also some poor effects was accrued. A number of players remained playing video games post COVID-19. The carried out survey is based on the social video games results on players' well-being and additionally on the effects of gamers' health and their sensible lifestyles. We are going to investigate the behavior of gamers engaged with video games during the COVID-19 lockdown and the video games affects on their well-being, the time they served in playing video games, and the consequential effect on their behavior and social and mental well-being. The results provide a start line for empirically grounded discussions on video games at some stage in the pandemic, their use, and potential outcomes. Different agegroups of players have been investigated. Most players are between 18 and 30 years. A number of the gamers during lock down played a few hours but most of players were males who spent most of their day playing video games. However, now the ratio of playing video games is reduced as examined with past circumstances. Roughly we can say that the condition as a whole is better, the reason why players enforced video games in their post COVID-19 practical life1. © 2023 IEEE.

关键词： Coronavirus

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Analysis of Blackhole Attack on RPL in Static, Mobile and Hybrid Network Environment

Comprehensive Analysis of Blackhole Attack on RPL in Static,...

引用

IEEE International Conference on Emerging & Sustainable Technologies for Power & ICT in a Developing Society (NIGERCON)

作者： Faisal Ali Garba Oluwakemi Oduwole Abubakar Balarabe Isa Abdulkadir Sarki Aliyu Rosemary Mrumun Dima Department of Computer Science Education Sa’Adatu Rimi University of Education Kano Nigeria Computer Science Programme National Mathematical Center Kwali-Sheda Abuja Nigeria Department of Software Engineering Federal University Dutsinma Katsina Nigeria

ISBN: (数字)9798331542559

ISBN: (纸本)9798331542566

The Routing Protocol for Low-Power and Lossy Networks (RPL), extensively applied in IoT applications, is susceptible to the Blackhole attack, wherein hostile nodes discard packets without transmission, resulting in considerable loss of network performance. This attack jeopardizes the reliability and usefulness of RPL-based networks, particularly in crucial applications like environmental monitoring and healthcare. Understanding the effects of Blackhole attacks in diverse network environments—static, mobile, and hybrid—is vital for creating personalized security solutions. Every scenario provides various issues, including the stability of static networks and the dynamic nature of mobile and hybrid setups, which determine the severity of the attack. This study performs a comparative investigation of the impact of Blackhole attacks on RPL-based networks in static, mobile, and hybrid scenarios. Simulations were done with varied node densities (10, 20, and 30 nodes) and different percentages of malicious nodes (10%, 20%, and 30%). Key performance indicators, such as Packet Delivery Ratio (PDR), Average End-to-End Delay (AE2ED), overhead packets, energy utilization, and throughput, were investigated across these conditions. The analysis suggests that static networks, however more robust to random attacks, are susceptible to disruption when Blackhole nodes interfere with important links. Mobile networks incur the highest performance deterioration due to frequent route changes and increasing energy demand. Hybrid networks, having both static and mobile nodes, demonstrate intermediate susceptibility but are more affected by mobile attackers. Overall, the statistics imply that mobility improves the impact of Blackhole attacks. The findings underscore the need for environment-specific security solutions in RPL-based networks. Static networks demand stronger detection systems for targeted attacks, while mobile and hybrid environments need adaptable, low-overhead security soluti

关键词： Energy consumption Network topology Throughput Propagation losses Hybrid power systems Stability analysis Routing protocols Topology Delays Reliability

来源：评论

学校读者我要写书评

暂无评论

Optimizing Gammatone Cepstral Coefficients for Gear Fault Detection

引用

IEEE Access 2025年 13卷 101711-101720页

作者： Zrar Kh Abdul Abdulbasit K. Al-Talabani Wisam Hazim Gwad Entisar Alkayal Halgurd S. Maghdid Safar Maghdid Asaad Computer Department College of Science Charmo University Sulaymaniyah Iraq Department of Software Engineering Faculty of Engineering Koya University Koya Iraq Department of Artificial Intelligence Engineering College of Engineering Alnoor University Ninavah Iraq Information Technology Department Faculty of Computing and Information Technology-Rabigh King Abdulaziz University Jeddah Saudi Arabia Department of Engineering Research Center Deanship of Research and Development Centers Koya University Erbil Governorate Iraq Department of Computer Engineering College of Engineering Knowledge University Erbil Iraq

Cepstral features, such as Gammatone Cepstral Coefficients (GTCC), have recently been applied in fault detection and diagnosis. However, GTCC was originally designed for speech feature extraction rather than fault detection, which limits its ability to capture relevant features for fault identification effectively. In this research, three key parameters of GTCC, namely, the number of coefficients, maximum frequency, and minimum frequency, are optimized using two metaheuristic algorithms: Fitness Dependent Optimizer (FDO) and Grey Wolf Optimization (GWO). These parameters are vital for enhancing the effectiveness of GTCC in extracting relevant features from the vibration signal for gear defect identification. Specifically, the maximum and minimum frequency values are critical for capturing the Gear Mesh Frequency (GMF), a common indicator of gear faults. Furthermore, optimizing the number of GTCC coefficients helps reduce model complexity. Experimental results demonstrate that optimizing GTCC parameters with GWO improves fault detection performance using an SVM classifier, achieving over 1% and 3% accuracy improvements on the PHM09 and DDS datasets, respectively.

关键词： Feature extraction Fault detection Optimization Gears Metaheuristics Mel frequency cepstral coefficient Mathematical models Filter banks Support vector machines Cepstrum

来源：评论

学校读者我要写书评

暂无评论

Explainable AI for Healthcare: An Approach Towards Interpretable Healthcare Models

Explainable AI for Healthcare: An Approach Towards Interpret...

引用

Information Technology (ACIT)

作者： Kanwal Zahoor Narmeen Zakaria Bawany Usman Ghani Department of Computer Science and Software Engineering Center for Computing Research Jinnah University for Women Institute of Business Administration Karachi Pakistan

Artificial intelligence (AI) along with deep learning techniques has become an integral part of almost all aspects of life. One of the domains significantly impacted by this technological revolution is healthcare. Deep learning-based AI systems assist clinicians and medical professionals in disease diagnosis, personalized treatment, and monitoring through wearables, among other applications. Despite its expedient integration in healthcare, the trustworthiness of deep learning models remains a concern, primarily due to a lack of understanding of their underlying processes. However, Explainable AI (XAI) offers explanations through various methods, including Local Interpretable Model-agnostic Explanations (LIME), Shapley Additive explanation (SHAP), and GRAD-CAM. XAI is utilized to enhance transparency, allowing users to understand and trust AI decisions. In this study, we present deep learning models for the classification of pneumonia disease in Chest X-ray Images followed by their explanations. Convolutional Neural Networks (CNNs) and other pre-trained models, including VGG16, MobileNetV3, and ResNet50, were used for classification of images as ‘normal’ or ‘pneumonia’. The VGG16 model, known for its exceptional image understanding capabilities, achieved the highest accuracy, with an impressive 93% score. Further, we used XAI techniques including SHAP, LIME, and Grad-CAM for explanation of models. LIME and Grad-CAM provided more accurate results than SHAP in our experiments. This approach was taken to evaluate the fairness and transparency of the model. The insights gained from XAI can be used to refine and improve machine learning models by identifying areas of weakness or misinterpretation which increases overall model robustness.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Cluster-Based Split Federated Learning Approach for Continuous User Authentication 7

Towards Cluster-Based Split Federated Learning Approach for ...

引用

7th Cyber Security in Networking Conference, CSNet 2023

作者： Wazzeh, Mohamad Arafeh, Mohamad Ould-Slimane, Hakima Talhi, Chamseddine Mourad, Azzam Otrok, Hadi Department of Software and It Engineering Montreal Canada Department of Mathematics and Computer Science Trois-Rivières Canada Cyber Security Systems and Applied Ai Research Center Lebanese American University Department of Csm Beirut Lebanon New York University Division of Science Abu Dhabi United Arab Emirates Khalifa University Department of Eecs Abu Dhabi United Arab Emirates

ISBN: (纸本)9798350342871

In today's rapidly evolving technological landscape, ensuring the security of systems requires continuous authentication over sessions and comprehensive access management during user interaction with a device. With the increasing use of smartphones and Internet of Things (IoT) devices, Split Learning (SL) and Federated Learning (FL) have emerged as promising technologies that can tackle the authentication problem while protecting the user's private data. The SL distributed technology enables users with limited resources to complete neural network model training with server assistance, lessening the computational burden from the client side. In addition, FL aims to combine knowledge between different nodes collaboratively. The privacy and security of the user's data are ensured in both approaches, as only the models' weights are shared with a server. This study employs a cluster-based approach using split learning and federated learning techniques to improve the efficiency and robustness of training Machine Learning (ML) models. We compare the approaches' performance to baseline methods and demonstrate their advantages using the UMDAA-02-FD face detection and MNIST datasets. Our findings show that combining both technologies achieves high accuracy in continuous authentication scenarios while maintaining user privacy. These results highlight the importance of SL and FL in cybersecurity, enabling continuous authentication and demonstrating their potential to revolutionize how we address security. © 2023 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning for Multi-Objective Resource Allocation in Multi-Platoon Cooperative Vehicular Networks

Deep Reinforcement Learning for Multi-Objective Resource All...

引用

作者： Xu, Yuanyuan Zhu, Kun Xu, Hu Ji, Jiequ Hohai University College of Computer and Information Nanjing211100 China Nanjing University of Aeronautics and Astronautics College of Computer Science and Technology Nanjing210016 China Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing211106 China University of Victoria Department of Electrical and Computer Engineering VictoriaV8P 5C2 Canada

Grouping vehicles into platoons is a promising cooperative driving scenario to enhance the traffic safety and capacity of future vehicular networks. However, fast changing channel conditions in multi-platoon vehicular networks cause tremendous uncertainty for resource allocation. In addition, the unprecedented proliferation of various emerging vehicle-to-infrastructure (V2I) applications may result in some service demands with conflicting quality of experience. In this paper, we formulate a multi-objective resource allocation problem, which maximizes the transmission success ratio of intra-platoon communications and the mean opinion score (MOS) of V2I communication links. To efficiently solve this multi-objective optimization problem, we resort to a deep reinforcement learning (DRL) framework. Specifically, we divide it into a set of scalar optimization subproblems based on the weighted sum approach and model each one as a partially observable stochastic game (P-OSG), where each platoon acts as an agent and the actions taken by all platoons correspond to the resource allocation solution. We further propose a contribution-based dual-clip proximal policy optimization (CD-PPO) algorithm to deal with each subproblem, which is a DRL algorithm based on the actor-critic framework. The network parameters of all subproblems are then optimized collaboratively by using the proposed training algorithm and the neighborhood parameter transfer strategy. The desired Pareto front is obtained when all subproblems are solved. Simulation results reveal that the proposed algorithm can outperform other algorithms in terms of the MOS and transmission success ratio. © 2002-2012 IEEE.

关键词： Reinforcement learning

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：