检索结果-内蒙古大学图书馆

Deep learning at the service of metaheuristics for solving numerical optimization problems

Neural Computing and Applications 2025年 1-36页

作者： Oyelade, Olaide N. Ezugwu, Absalom E. Saha, Apu K. Thieu, Nguyen V. Gandomi, Amir H. School of Electronics Electrical Engineering and Computer Science Queen’s University Belfast Bangladesh Unit for Data Science and Computing North-West University 11 Hoffman Street Potchefstroom2520 United States Department of Mathematics National Institute of Technology Agartala Tripura Agartala799046 India Faculty of Computer Science Phenikaa University Yen Nghia Hanoi Ha Dong12116 Viet Nam Department of Engineering and Information Technology University of Technology Sydney UltimoNSW2007 Australia

Integrating deep learning methods into metaheuristic algorithms has gained attention for addressing design-related issues and enhancing performance. The primary objective is to improve solution quality and convergence speed within solution search spaces. This study investigates the use of deep learning methods as a generative model to learn historical content, including global best and worst solutions, solution sequences, function evaluation patterns, solution space characteristics, population modification trajectories, and movement between local and global search processes. An LSTM-based architecture is trained on dynamic optimization data collected during the metaheuristic optimization process. The trained model generates an initial solution space and is integrated into the optimization algorithms to intelligently monitor the search process during exploration and exploitation phases. The proposed deep learning-based methods are evaluated on 55 benchmark functions of varying complexities, including CEC 2017 and compared with 13 biology-based, evolution-based, and swarm-based metaheuristic algorithms. Experimental results demonstrate that all the deep learning-based optimization algorithms achieve high-quality solutions, faster convergence rates, and significant performance improvements. These findings highlight the critical role of deep learning in addressing design issues, enhancing solution quality, trajectory, and performance speed in metaheuristic algorithms. © The Author(s) 2025.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

High-Performance and Intelligent Digital Active Sensor System for Epilepsy Information Acquisition and Classification

引用

IEEE Sensors Journal 2025年第9期25卷 16392-16410页

作者： Sun, Jianhui Chen, Xiaodong Yao, Wenlong Liu, Guozhu Li, Tianlai Liu, Zhenpeng Jiang, Zekun Shandong Normal University School of Physics and Electronics Jinan250358 China Key Laboratory of Transducer Technology Institute of Electronics Chinese Academy of Sciences Beijing100190 China 58th Research Institute of China Electronics Technology Group Corporation Chinese Academy of Sciences Wuxi214035 China Southeast University School of Electronic Science and Engineering Nanjing210096 China Shandong Normal University Informatization Office Jinan250358 China Sichuan University College of Computer Science Chengdu610041 China

In this article, we present a performance-enhanced digital active sensor system consisting of a quantum-enhanced active sensor (that frontend micro-electro-mechanical-systems (MEMS) electrode coupled with readout circuits) and an artificial intelligence (AI) empowered engine, which is used for detecting dual-mode neural signals sensitively and classifying about epilepsy disease accurately, respectively. First, digital active sensors with improved sensitivity and reliability include: 1) the active sensor’s bare microelectrode array (MEA) manufactured from the perspective of quantum mechanics;2) the high-resolution active sensor’s information quantizer for weak neural electrophysiological signal quantization with noise depression techniques;and 3) transmitting power of the radio transmitter reduces by employing bitstream compression with fast rolling-in hash table method. Second, structure and subsystem re-synthesizable AI engine with flexible and scalability abilities. For bare microelectrodes, on the one side, an optimized stack Si3N4/SiO2 dielectric thin film of the bare microelectrode can prevent leakage caused by Fowler-Nordheim (FN) tunneling and stress-induced leakage-current (SILC) effects, also solve the microelectrode peeling problem caused by mechanical stress. In addition, nanomaterial Pt with nanostructure modified on the surface of Pt/Ti alloy of the bare microelectrode for reconstructing local Fermi level, increasing charge injection, and boosting electrocatalytic with continuous-descent-energy-step (CDES) approach. For the neurotransmitter microelectrode manufactured based on the bare microelectrode, the tested linear range is 0.998° and sensitivity is 0.0072 pA/ µm2. Energy consumption of the entire digital active sensor is less than 70 mW under a 1.8-V power and electrophysiological spike resolution is 0.93 µV. The recognition accuracy of the resynthesizable AI engine arrives at 99.74% using a random forest model, which is appropriate for epilepsy di

关键词： Microelectrodes

来源：评论

学校读者我要写书评

暂无评论

NumScout: Unveiling Numerical Defects in Smart Contracts using LLM-Pruning Symbolic Execution

引用

IEEE Transactions on Software Engineering 2025年第05期51卷 1538-1553页

作者： Chen, Jiachi Shao, Zhenzhe Yang, Shuo Shen, Yiming Wang, Yanlin Chen, Ting Shan, Zhenyu Zheng, Zibin Hangzhou Normal University Intelligent Transportation and Information Security Laboratory Hangzhou311121 China Sun Yat-sen University School of Software Engineering Zhuhai519082 China University of Electronic Science and Technology of China School of Computer Science and Engineering School of Cyber Security Chengdu611731 China Kashi Institute of Electronics and Information Industry Kashi844000 China

In recent years, the Ethereum platform has witnessed a proliferation of smart contracts, accompanied by exponential growth in total value locked (TVL). High-TVL smart contracts often require complex numerical computations, particularly in mathematical financial models used by many decentralized applications (DApps). Improper calculations can introduce numerical defects, posing potential security risks. Existing research primarily focuses on traditional numerical defects like integer overflow, and there is currently a lack of systematic research and effective detection methods targeting new types of numerical defects. In this paper, we identify five new types of numerical defects through the analysis of 1,199 audit reports by utilizing the open card method. Each defect is defined and illustrated with a code example to highlight its features and potential consequences. We also propose NumScout, a symbolic execution-based tool designed to detect these five defects. Specifically, the tool combines information from source code and bytecode, analyzing key operations such as comparisons and transfers, to effectively locate defects and report them based on predefined detection patterns. Furthermore, NumScout uses a large language model (LLM) to prune functions which are unrelated to numerical operations. This step allows symbolic execution to quickly enter the target function and improve runtime speed by 28.4%. We run NumScout on 6,617 real-world contracts and evaluated its performance based on manually labeled results. We find that 1,774 contracts contained at least one of the five defects, and the tool achieved an overall precision of 89.7%. © 2025 IEEE.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Language-Guided Semantic Alignment for Point Cloud-Based 3D Scene Graph Generation

Language-Guided Semantic Alignment for Point Cloud-Based 3D ...

引用

2025 International Conference on electronics, Information, and Communication, ICEIC 2025

作者： Lee, Suhyeon Jeong, Jinwoo Jeong, Taehyun Kim, Sungjei Intelligent Image Processing Research Center Korea Electronics Technology Institute Seongnam Korea Republic of School of Computer Science and Engineering Korea University of Technology and Education Cheonan Korea Republic of

ISBN: (纸本)9798331510756

In this paper, we propose a novel model called LAP, Language-guided semantic Alignment for Point cloud-based 3D semantic scene graph generation. Point cloud-based 3D scene graph generation is the task of predicting the category of each instance present in a 3D point cloud scene and the relationships between them. Our model leverages text embeddings for objects, predicates, and triplets during network training, enabling the network to generate reliable semantic scene graphs. To this end, we introduce three loss functions that leverage text embed dings to semantically align the features for scene graph prediction. Comparative experiments with previous methods on the 3DSSG dataset demonstrate the effectiveness of the LAP model. © 2025 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Secure beamforming designs for maximizing secrecy sum rate in MISO-NOMA networks

引用

Digital Communications and Networks 2025年第1期11卷 83-91页

作者： Yanbo Zhang Zheng Yang Jingjing Cui Xianfu Lei Yi Wu Jun Zhang Chao Fang Zhiguo Ding College of Photonic and Electronic Engineering Fujian Normal UniversityFuzhou 350007China School of Electronics and Computer Science University of SouthamptonSouthampton SO171BJUK Institute of Mobile Communications Southwest Jiaotong UniversityChengdu 610031China Jiangsu Key Laboratory of Wireless Communications Nanjing University of Posts and TelecommunicationsNanjing 210003China Faculty of Information Technology Beijing University of TechnologyBeijingChina School of Electrical and Electronics Engineering The University of ManchesterManchester M139PLUK

In this paper,the application of Non-Orthogonal Multiple Access(NOMA)is investigated in a multiple-input single-output network consisting of multiple legitimate users and a potential *** support secure transmissions from legitimate users,two NOMA Secrecy Sum Rate Transmit Beam Forming(NOMA-SSR-TBF)schemes are proposed to maximise the SSR of a Base Station(BS)with sufficient and insufficient transmit *** BS with sufficient transmit power,an artificial jamming beamforming design scheme is proposed to disrupt the potential eavesdropping without impacting the legitimate *** addition,for BS with insufficient transmit power,a modified successive interference cancellation decoding sequence is used to reduce the impact of artificial jamming on legitimate *** specifically,iterative algorithm for the successive convex approximation are provided to jointly optimise the vectors of transmit beamforming and artificial *** results demonstrate that the proposed NOMA-SSR-TBF schemes outperforms the existing works,such as the maximized artificial jamming power scheme,the maximized artificial jamming power scheme with artificial jamming beamforming design and maximized secrecy sum rate scheme without artificial jamming beamforming design.

关键词： NOMA Secure transmission Artificial jamming Beamforming design

来源：评论

学校读者我要写书评

暂无评论

An Iterative PRISMA Review of GAN Models for Image Processing, Medical Diagnosis, and Network Security

引用

computers, Materials & Continua 2025年第2期82卷 1757-1810页

作者： Uddagiri Sirisha Chanumolu Kiran Kumar Sujatha Canavoy Narahari Parvathaneni Naga Srinivasu Department of Computer Science and Engineering Prasad V Potluri Siddhartha Institute of TechnologyVijayawada520007India Department of Computer Science and Engineering Koneru Lakshmaiah Education FoundationVaddeswaram522302India Department of Electronics and Communication Engineering Sreenidhi Institute of Science and TechnologyHyderabad501301India Amrita School of Computing Amrita Vishwa VidyapeethamAmaravati522503India Department of Teleinformatics Engineering Federal University of CearáFortaleza60455-970Brazil Inti International University Nilai71800Malaysia

The growing spectrum of Generative Adversarial Network (GAN) applications in medical imaging, cyber security, data augmentation, and the field of remote sensing tasks necessitate a sharp spike in the criticality of review of Generative Adversarial Networks. Earlier reviews that targeted reviewing certain architecture of the GAN or emphasizing a specific application-oriented area have done so in a narrow spirit and lacked the systematic comparative analysis of the models’ performance metrics. Numerous reviews do not apply standardized frameworks, showing gaps in the efficiency evaluation of GANs, training stability, and suitability for specific tasks. In this work, a systemic review of GAN models using the PRISMA framework is developed in detail to fill the gap by structurally evaluating GAN architectures. A wide variety of GAN models have been discussed in this review, starting from the basic Conditional GAN, Wasserstein GAN, and Deep Convolutional GAN, and have gone down to many specialized models, such as EVAGAN, FCGAN, and SIF-GAN, for different applications across various domains like fault diagnosis, network security, medical imaging, and image segmentation. The PRISMA methodology systematically filters relevant studies by inclusion and exclusion criteria to ensure transparency and replicability in the review process. Hence, all models are assessed relative to specific performance metrics such as accuracy, stability, and computational efficiency. There are multiple benefits to using the PRISMA approach in this setup. Not only does this help in finding optimal models suitable for various applications, but it also provides an explicit framework for comparing GAN performance. In addition to this, diverse types of GAN are included to ensure a comprehensive view of the state-of-the-art techniques. This work is essential not only in terms of its result but also because it guides the direction of future research by pinpointing which types of applications require some

关键词： GAN CGAN WGAN DCGAN image analysis

来源：评论

学校读者我要写书评

暂无评论

Extracting Facial Features to Detect Deepfake Videos Using Machine Learning

引用

International Journal of Advanced computer science and Applications 2025年第4期16卷 834-842页

作者： Aslam, Ayesha Mir, Jamaluddin Zaman, Gohar Rahman, Atta Salam, Asiya Abdus Ali, Farhan Alhiyafi, Jamal Bakry, Aghiad Gul, Mustafa Jamal Gollapalli, Mohammed Mahmud, Maqsood Department of Computer Science Abbottabad University of Science and Technology Havelian Pakistan Department of Computer Science College of Computer Science and Information Technology Imam Abdulrahman Bin Faisal University P.O. Box 1982 Dammam31441 Saudi Arabia Department of Computer Information Systems College of Computer Science and Information Technology Imam Abdulrahman Bin Faisal University P.O. Box 1982 Dammam31441 Saudi Arabia College of Electronics and Information Engineering Shenzhen University Shenzhen China Department of Computer Science Kettering University FlintMI United States Department of Business Administration University of York York YO10 5DD Heslington United Kingdom Department of Information Technology & Engineering Sydney Met SydneyNSW2000 Australia School of Computing Ulster University Northern Ireland Belfast United Kingdom

Generative adversarial networks (GANs) have gained popularity for their ability to synthesize images from random inputs in deep learning models. One of the notable applications of this technology is the creation of realistic videos known as deepfakes, which have been misused on social media platforms. The difficulty lies in distinguishing these fake videos from real ones with the naked eye, leading to significant concerns. This study proposes a supervised machine learning approach to effectively differentiate between real and counterfeit videos by detecting visual artifacts. To achieve this, two facial features are extracted: eye blinking and nose position, utilizing landmark detection techniques. Both features were trained on supervised machine learning classifiers and evaluated using the publicly available UADFV and Celeb-DF deepfake datasets. The experiments successfully demonstrate that the proposed method achieves a promising and superior performance, with an area under the curve (AUC) of 97% for deepfake detection in contrast to state-of-the-art methods investigating the same datasets. © (2025), (science and Information Organization). All rights reserved.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Augmented Evolutionary Strategies for Intelligent Global Optimization

引用

IEEE Access 2025年 13卷 70138-70178页

作者： Ezugwu, Absalom El-Shamir Oyelade, Olaide Nathaniel Agushaka, Jeffrey Ovre Saha, Apu Kumar Unit for Data Science and Computing North-West University Potchefstroom2520 South Africa School of Electronics Electrical Engineering and Computer Science Queen’s University BelfastBT9 5BN United Kingdom Department of Computer Science Federal University of Lafia Lafia950101 Nigeria Department of Mathematics National Institute of Technology Agartala Tripura Agartala799046 India

This paper introduces the Susceptible-Infected-Removed Optimizer (SIRO), a novel learned heuristic inspired by biological systems and deep learning techniques. SIRO models its search process after the SIR epidemiological compartmental model, predicting the susceptibility, infection, and recovery dynamics of solutions. SIRO integrates deep learning into its initialization and parameter setting to enhance its efficiency, enabling intelligent and adaptive behavior. This hybridization improves solution quality, accelerates convergence, enhances robustness, and reduces computational costs. The algorithm’s performance was evaluated using CEC 2017 benchmark functions, demonstrating superior results in hybrid functions (C1-C28) despite moderate performance on traditional CEC1-CEC14 functions. Friedman’s test ranked SIRO 4th overall, with SSA as the top-performing algorithm. Additionally, SIRO was tested on real-world optimization problems, including mechanical engineering design, hyperparameter tuning, and feature selection for medical image classification. In the classification task, SIRO-enhanced CNN achieved an accuracy of 0.86 at the 5th epoch, outperforming CNN (0.66), CNN-GA (0.76), and CNN-WOA (0.75). Furthermore, SIRO reported a precision of 0.96, recall of 1.0, and F1-score of 0.98, highlighting its effectiveness. These results validate the benefits of integrating a learning mechanism into SIRO, yielding superior precision, computational efficiency, and performance over conventional optimization approaches. © 2013 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing Skin Cancer Detection with Multimodal Data Integration: A Combined Approach Using Images and Clinical Notes

引用

SN computer science 2025年第1期6卷 1-14页

作者： Chakkarapani, V. Poornapushpakala, S. Suresh, S. School of Electrical and Electronics Sathyabama Institute of Science and Technology (Deemed to be University) Jeppiaar Nagar Rajiv Gandhi Salai Tamilnadu Chennai 600119 India Department of Computer Applications National Institute of Technology Haryana Kurukshetra 136119 India

Skin cancer is one of the most common cancers worldwide where early detection is crucial for effectively diagnosing and treatment. Traditional diagnostic methods largely depend on dermoscopic or skin lesion image analysis. However, they are limited by the clinician’s expertise and primarily rely on image data. While most of the studies focus on improving the diagnostic accuracy using the skin lesion images, this study addresses the limitations of using just the image data and proposes a novel approach that enhances skin cancer detection by combining skin lesion images with patients’ clinical notes. The methodology proposes a multi-modal approach that combines the images data with doctor’s clinical notes to enhance early diagnosis and accurate detection of skin cancer. Clinical notes are synthetically generated using two distinct multimodal large language models (LLMs), corresponding to each patient’s skin cancer images, encapsulating both visual and textual data representative of real-world scenarios. The generated clinical notes are validated through a dual-method approach involving cross-evaluation and consensus scoring, utilizing metrics such as the BLEU score, ROUGE score, Overlap Coefficient, and Jaccard index. Furthermore, this paper compares the performance of a model built using skin lesion images with four other models built using skin lesion images and synthetically generated clinical notes. The experimental results demonstrates a significant improvement in all classification metrics compared to single-modality models, achieving an accuracy of 99.51%, precision of 96.19%, recall of 97.95%and f1-score of 97.03% with the multimodal ALBEF model that uses GPT-4-turbo for clinical notes generation. The results show a significant improvement in performance with the multi modal approach for medical diagnosis. Also, this research sets a foundational framework that can be leveraged for many potential healthcare diagnoses applications. © The Author(s), under exclusi

关键词： Dermatology Generative AI Large language models Multi-modal AI Skin cancer detection Synthetic data generation

来源：评论

学校读者我要写书评

暂无评论

Federated learning-based hybrid convolutional recurrent neural network for multi-class intrusion detection in IoT networks

引用

Discover Internet of Things 2025年第1期5卷 1-22页

作者： Selvam, Prabu Karthikeyan, P. Manochitra, S. Sujith, A.V.L.N. Ganesan, T. Ayyasamy, Rajaram Shuaib, Mohammed Alam, Shadab Rajendran, A. School of Computing SRM Institute of Science and Technology Tiruchirappalli621105 India Department of Electronics and Communication Engineering Velammal College of Engineering and Technology Tamil Nadu Madurai625009 India Department of Computer Science Engineering St Joseph’s College of Engineering Tamil Nadu Chennai600119 India School of Engineering Malla Reddy University Telangana Hyderabad500043 India Department of Computer Science Engineering Koneru Lakshmaiah Education Foundation Andhra Pradesh Vaddeswaram522302 India Department of Electronics and Communication Engineering E.G.S Pillay Engineering College Tamil Nadu Nagapattinam611002 India Department of Computer Science College of Engineering and Computer Science Jazan University Jazan Saudi Arabia Department of ECE Karpagam College of Engineering Tamil Nadu Othakkal Mandapam India Chitkara University Institute of Engineering and Technology Chitkara University Punjab Rajpura India

As Internet of Things (IoT) devices leads an significant challenges in securing the systems from cyber-attacks in large-scale IoT networks. Traditional methods faces struggle to precisely detecting the complex intrusion patterns due to the heterogeneity and distributed IoT devices. There is a pressing need for scalable, real-time, and privacy-preserving solutions to secure IoT and IIoT systems. For multi-class intrusion detection in IoT networks, this study proposed an novel hybrid convolutional recurrent neural network (CRNN) model based on federated learning. By integrating federated learning, the proposed approach ensures privacy by enabling decentralized model training among IoT devices without exchanging sensitive data. The hybrid CRNN model improves detection accuracy for a variety of attack types in IoT networks by combining Recurrent Neural Networks for sequential pattern recognition and Convolutional Neural Networks for feature extraction from raw data. The proposed model classifies incoming network traffic into different attack classes. The experimental results conducted on the Edge-IIoT dataset reveal that the proposed model has attained the detection accuracy of 98.93% with balanced precision and recall for all the attack types and low-support classes such as SQL Injection and Man-in-the-Middle. This balance improves model’s ability to reduce both false negatives and false positives compared with conventional approaches. Together with real-time detection, our approach excels in a privacy-preserving manner, providing a practical, scalable solution for securing complex IoT networks, and addressing critical gaps in current IDS frameworks. © The Author(s) 2025.

关键词： Federated 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：