检索结果-内蒙古大学图书馆

Arabic Music Genre Classification Using Deep Convolutional Neural Networks (CNNs)

computers, Materials & Continua 2022年第9期72卷 5443-5458页

作者： Laiali Almazaydeh Saleh Atiewi Arar Al Tawil Khaled Elleithy Department of Software Engineering Al-Hussein Bin Talal UniversityMa’an71111Jordan Department of Computer Science Al-Hussein Bin Talal UniversityMa’an71111Jordan King Abdullah II School of Information Technology The University of JordanAmmanJordan Department of Computer Science and Engineering University of BridgeportBridgeportCT06604USA

Genres are one of the key features that categorize music based on specific series of ***,the Arabic music content on the web is poorly defined into its genres,making the automatic classification of Arabic audio genres *** this reason,in this research,our objective is first to construct a well-annotated dataset of five of the most well-known Arabic music genres,which are:Eastern Takht,Rai,Muwashshah,the poem,and Mawwal,and finally present a comprehensive empirical comparison of deep Convolutional Neural Networks(CNNs)architectures on Arabic music genres *** this work,to utilize CNNs to develop a practical classification system,the audio data is transformed into a visual representation(spectrogram)using Short Time Fast Fourier Transformation(STFT),then several audio features are extracted using Mel Frequency Cepstral Coefficients(MFCC).Performance evaluation of classifiers is measured with the accuracy score,time to build,and Matthew’s correlation coefficient(MCC).The concluded results demonstrated that AlexNet is considered among the topperforming five CNNs classifiers studied:LeNet5,AlexNet,VGG,ResNet-50,and LSTM-CNN,with an overall accuracy of 96%.

关键词： CNN MFCC spectrogram STFT arabic music genres

来源：评论

学校读者我要写书评

暂无评论

Stochastic Q-learning for large discrete action spaces 24

Stochastic Q-learning for large discrete action spaces

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Fares Fourati Vaneet Aggarwal Mohamed-Slim Alouini Computer Electrical and Mathematical Science and Engineering Division King Abdullah University of Science and Technology (KAUST) Thuwal KSA School of Industrial Engineering Purdue University West Lafayette IN

In complex environments with large discrete action spaces, effective decision-making is critical in reinforcement learning (RL). Despite the widespread use of value-based RL approaches like Q-learning, they come with a computational burden, necessitating the maximization of a value function over all actions in each iteration. This burden becomes particularly challenging when addressing large-scale problems and using deep neural networks as function approximators. In this paper, we present stochastic value-based RL approaches which, in each iteration, as opposed to optimizing over the entire set of n actions, only consider a variable stochastic set of a sublinear number of actions, possibly as small as O(log(n)). The presented stochastic value-based RL methods include, among others, Stochastic Q-learning, StochDQN, and StochDDQN, all of which integrate this stochastic approach for both value-function updates and action selection. The theoretical convergence of Stochastic Q-learning is established, while an analysis of stochastic maximization is provided. Moreover, through empirical validation, we illustrate that the various proposed approaches outperform the baseline methods across diverse environments, including different control problems, achieving near-optimal average returns in significantly reduced time.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient GPU-based Large MIMO Detection Algorithm for Next-Generation Communication Systems

Efficient GPU-based Large MIMO Detection Algorithm for Next-...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： A. Dabah Z. Rezki H. Ltaief D. Keyes M.-S. Alouini Computer Electrical and Mathematical Science and Engineering King Abdullah University of Science and Technology University of California Santa Cruz USA

Low latency and high throughput are critical features for 5G mobile communication systems and beyond, in which the support of large MIMO is essential. Signal detection in large Multiple-Input Multiple-Output (MIMO) is a paramount component of a communication system since its performance in terms of latency, error rate, and achieved throughput depends on it. In this paper, we demonstrate the ability of our proposed massively parallel non-linear detection approach to support a large number of antennas and sustain high throughput at the extreme low latency of next-generation mobile communication systems. Our proposed method operates on a search tree that models all possible combinations of the transmitted signal. It selects coefficients from different levels and navigates the tree toward the Maximum Likelihood (ML) solution. To maintain the low latency requirement, we leverage the significant computational power of the Graphics Processing Unit (GPU) by expressing operations in terms of matrix-matrix multiplications. The obtained results show the ability of our non-linear detection approach to deal with up to 120 antennas with one-millisecond latency while satisfying good error rate performance at a practical signal-to-noise ratio (SNR).

关键词：

来源：评论

学校读者我要写书评

暂无评论

Human Pose Estimation and Object Interaction for Sports Behaviour

引用

computers, Materials & Continua 2022年第7期72卷 1-18页

作者： Ayesha Arif Yazeed Yasin Ghadi Mohammed Alarfaj Ahmad Jalal Shaharyar Kamal Dong-Seong Kim Department of Computer Science Air UniversityIslamabad44000Pakistan Department of Computer Science and Software Engineering Al Ain UniversityAl Ain15551UAE Department of Electrical Engineering College of EngineeringKing Faisal UniversityAl-AhsaSaudi Arabia Department of IT Convergence Engineering Kumoh National Institute of TechnologyGumiKorea

In the new era of technology,daily human activities are becoming more challenging in terms of monitoring complex scenes and *** understand the scenes and activities from human life logs,human-object interaction(HOI)is important in terms of visual relationship detection and human pose *** understanding and interaction recognition between human and object along with the pose estimation and interaction modeling have been *** existing algorithms and feature extraction procedures are complicated including accurate detection of rare human postures,occluded regions,and unsatisfactory detection of objects,especially small-sized *** existing HOI detection techniques are instancecentric(object-based)where interaction is predicted between all the *** estimation depends on appearance features and spatial ***,we propose a novel approach to demonstrate that the appearance features alone are not sufficient to predict the ***,we detect the human body parts by using the Gaussian Matric Model(GMM)followed by object detection using *** predict the interaction points which directly classify the interaction and pair them with densely predicted HOI vectors by using the interaction *** interactions are linked with the human and object to predict the *** experiments have been performed on two benchmark HOI datasets demonstrating the proposed approach.

关键词： Human object interaction human pose estimation object detection sports estimation sports prediction

来源：评论

学校读者我要写书评

暂无评论

Discrete Anomalous Hall Resistance-based Quantized Convolutional Neural Network

Discrete Anomalous Hall Resistance-based Quantized Convoluti...

引用

Magnetic Conference - Short Papers (INTERMAG Short Papers), IEEE International

作者： Aijaz H. Lone Xuecui Zou Hanrui Li Nazek El-Atab Hossein Fariborzi Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Saudi Arabia

Spintronic devices have shown promise for energy-efficient storage and neuromorphic computing. In this abstract, we present the realization of a spintronic device exhibiting discrete anomalous Hall resistance states. We attribute this discrete resistance behavior to the magnetic domain wall pinning and depinning and gradual switching of different magnetic layers. The number of resistance states is a function of the temperature. Furthermore, this discrete resistance behavior of the device allows us to employ these resistance states as weights in a quantized convolutional neural network. The network is trained and tested on the CIFAR-10 data set and the system achieves an accuracy of around 86.95%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-UAV Placement for Integrated Access and Backhauling Using LLM-Driven Optimization

Multi-UAV Placement for Integrated Access and Backhauling Us...

引用

2025 IEEE Wireless Communications and Networking Conference, WCNC 2025

作者： Wang, Yuhui Farooq, Junaid Ghazzai, Hakim Setti, Gianluca University of Michigan-Dearborn Department of Electrical and Computer Engineering DearbornMI48126 United States King Abdullah University of Science & Technology Computer Electrical and Mathematical Sciences Division Saudi Arabia

ISBN: (纸本)9798350368369

Unmanned aerial vehicles (UAVs) can enhance wireless access by dynamically positioning themselves closer to users while maintaining a backhaul connection to cellular base stations. In scenarios where users are geographically dispersed, multiple UAVs can be orchestrated to establish multi-hop integrated access and backhaul (IAB) connections. Traditionally, finding optimal UAV placement has required computationally demanding methods, such as combinatorial optimization or reinforcement learning, which are often impractical for real-time applications due to their complexity and training requirements. Additionally, UAV operators may lack the capability to solve complex optimization problems during live operations. This paper presents a novel framework that leverages large language models (LLMs) for optimizing the placement of multiple UAVs through iterative structured prompting. The proposed method achieves near-optimal solutions in significantly fewer iterations compared to traditional methods, making it suitable for real-time deployment without extensive mathematical modeling. Simulation result demonstrate that the proposed method achieves scores over 82% of the theoretical optimal solution while reducing computational time from hours to minutes compared to the baseline deep Q network approach, ensuring robust network connectivity and service quality. The LLM-driven framework simplifies problem-solving for UAV network operators, paving the way for its application in more complex real-world scenarios. © 2025 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Federated combinatorial multi-agent multi-armed bandits 24

Federated combinatorial multi-agent multi-armed bandits

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Fares Fourati Mohamed-Slim Alouini Vaneet Aggarwal Computer Electrical and Mathematical Science and Engineering Division King Abdullah University of Science and Technology (KAUST) Thuwal KSA School of Industrial Engineering Purdue University West Lafayette IN

This paper introduces a federated learning framework tailored for online combinatorial optimization with bandit feedback. In this setting, agents select subsets of arms, observe noisy rewards for these subsets without accessing individual arm information, and can cooperate and share information at specific intervals. Our framework transforms any offline resilient single-agent (α - ε)-approximation algorithm--having a complexity of Õ(ψ/εβ), where the logarithm is omitted, for some function ψ and constant β--into an online multi-agent algorithm with m communicating agents and an a-regret of no more than Õ(m-1/3+β ψ1/3+βT2+β/3+β). Our approach not only eliminates the ε approximation error but also ensures sublinear growth with respect to the time horizon T and demonstrates a linear speedup with an increasing number of communicating agents. Additionally, the algorithm is notably communication-efficient, requiring only a sublinear number of communication rounds, quantified as Õ(ψTβ/β+1). Furthermore, the framework has been successfully applied to online stochastic submodular maximization using various offline algorithms, yielding the first results for both singleagent and multi-agent settings and recovering specialized single-agent theoretical guarantees. We empirically validate our approach to a stochastic data summarization problem, illustrating the effectiveness of the proposed framework, even in single-agent scenarios.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Epilepsy seizure detection based on EEG QuPWM features and Logistic Regression

引用

IFAC-PapersOnLine 2024年第24期58卷 356-361页

作者： María de los Ángeles Gómez Castillo Taous-Meriem Laleg Kirati Bio-engineering Program King Abdullah University of Science and Technology (KAUST) Thuwal KSA National Institute for Research in Digital Science and Technology Paris-Saclay France.

Epilepsy is one of the most common neurological diseases that affects around 50 million people worldwide of all ages according to the World Health Organization. A standard test widely used to detect abnormalities in the electrical activity of the brain, such as seizures is the electroencephalogram (EEG). EEG is commonly analyzed and examined manually by expert reviewers to detect epileptic activity. A method that is time-consuming and tedious, may lead to errors and inconsistencies, especially for long EEG recordings. Therefore, automatic seizure detection is meaningful for the accurate diagnosis and monitoring of epilepsy, to initiate the proper treatment, and subsequently reduce the risk of future seizures. This study proposes a computer-aided diagnosis for automatic epileptic seizure detection from EEG data. In this study, an adaptation of the quantization-based position weight matrix (QuPWM) is proposed to extract features from the frequency domain of the framed EEG signal, these features are fed to a Logistic Regression(LR) to detect epileptic seizures. The performance has been evaluated on the university of Bonn EEG dataset and compared to the state-of-the-art methods. The proposed model resulted in satisfactory classification accuracy, sensitivity, and specificity rates.

关键词： EEG Position Weight Matrix Logistic Regression seizure detection

来源：评论

学校读者我要写书评

暂无评论

Ontological analysis of collection improves classification of cardinality phenotypes 13

Ontological analysis of collection improves classification o...

引用

13th International Conference on Biomedical Ontology, ICBO 2022

作者： Alghamdi, Sarah Hoehndorf, Robert Computational Bioscience Research Center Computer Electrical & Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology 4700 KAUST Thuwal23955 Saudi Arabia

Phenotype ontologies formally characterize and classify phenotypes. We analyze the phenotype classes related to the presence of increased or decreased amount of entities and identify potentially misleading inferences as a consequence of the axioms used to formalize these classes. We propose an ontology design pattern to reformulate these cardinality-related phenotypes and show how these can lead to a clearer classification and higher expressivity. Furthermore, reformulating these phenotype classes in MP and HP improves predictive performance in the task of identifying gene–disease associations through semantic similarity. © 2022 Copyright for this paper by its authors.

关键词： Latent semantic analysis

来源：评论

学校读者我要写书评

暂无评论

Analysis of 3D localization in underwater optical wireless networks with uncertain anchor positions

引用

science China(Information sciences) 2020年第10期63卷 174-181页

作者： Nasir SAEED Abdulkadir CELIK Mohamed-Slim ALOUINI Tareq Y.AL-NAFFOURI Computer ElectricalMathematical Sciences & Engineering (CEMSE) DivisionKing Abdullah University of Science and Technology (KAUST)

Localization accuracy is of paramount importance for the proper operation of underwater optical wireless sensor networks(UOWSNs). However, underwater localization is prone to hostile environmental impediments such as drifts owing to the surface and deep currents. These cause uncertainty in the deployed anchor node positions and pose daunting challenges to achieve accurate location estimations. Therefore, this paper analyzes the performance of three-dimensional(3D) localization for UOWSNs and derives a closedform expression for the Cramer Rao lower bound(CRLB) by using time of arrival(ToA) and angle of arrival(AoA) measurements under the presence of uncertainty in anchor node positions. Numerical results validate the analytical findings by comparing the localization accuracy in scenarios with and without anchor nodes position uncertainty. Results are also compared with the linear least square(LLS) method and weighted LLS(WLLS) method.

关键词： underwater optical wireless sensor networks localization three-dimensional uncertainty

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：