检索结果-内蒙古大学图书馆

Open and Closed Source Models for LLM-Generated Metaheuristics Solving engineering Optimization Problem 28th

Open and Closed Source Models for LLM-Generated Metaheuris...

28th European Conference on Applications of Evolutionary Computation, EvoApplications 2025, held as part of EvoStar 2025

作者： Senkerik, Roman Viktorin, Adam Kadavy, Tomas Kovac, Jozef Janku, Peter Pekar, Libor Guzowski, Hubert Smolka, Maciej Byrski, Aleksander Pluhacek, Michal A.I.Lab Faculty of Applied Informatics Tomas Bata University Zlin Czech Republic Department of Automation and Control Engineering Faculty of Applied Informatics Tomas Bata University Zlin Czech Republic Faculty of Computer Science AGH University of Krakow Krakow Poland Center of Excellence in Artificial Intelligence AGH University of Krakow Krakow Poland

ISBN: (纸本)9783031900648

This paper explores the applicability of generative AI (genAI), specifically Large Language Models (LLMs), for the automatic generation and configuration of metaheuristic algorithms to address a real-world engineering problem: the optimal parameter estimation of time-delay systems in interconnected heating-cooling loops. The study introduces a pioneering workflow and iterative architecture with feedback within the emerging field of genAI-driven optimization for real optimization problems, eliminating the need for manually crafted or modified algorithms. This automated system empowers domain experts in engineering to solve complex optimization problems with minimal knowledge of optimization algorithms, lowering the barrier to entry for sophisticated algorithm use. We demonstrate how LLMs can generate effective optimizers under conditions like connstrained optimization problems where the solution lies near the boundaries of the search space. Four state-of-the-art LLMs (closed and open-sourced) have been selected for experiments. These are GPT-4o, GPT-4o mini, Claude Sonnet 3.5 and Llama 3.1. All studied LLMs generated metaheuristics that outperformed the initialization baseline optimization method (Random Search and CMA-ES). Notably, the Claude Sonnet 3.5 model generated a metaheuristic with the best mean results, almost matching the performance of the tuned state-of-the-art DISH algorithm, as an example of adaptive Differential Evolution. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Time of arrival

来源：评论

学校读者我要写书评

暂无评论

Enhanced Speech Emotion and Gender Recognition Using Hybrid CNN-LSTM Models

Enhanced Speech Emotion and Gender Recognition Using Hybrid ...

引用

Sentiment Analysis and Deep Learning (ICSADL), International Conference on

作者： S Priyadharshini G. Naveen Sundar D. Narmadha Senthil Kumar Thangavel Adeyemi Abel Ajibesin Selvanayaki Kolandapalayam Shanmugam Department of Computer Science and Engineering Karunya Institute of Technology and Sciences Coimbatore India Department of Computer Science and Engineering Amrita School of Computing Coimbatore India Department of Information Technology Faculty of Informatics and Design Cape Peninsula University of Technology South Africa Department of Mathematics and Computer Science Ashland University USA

ISBN: (数字)9798331523923

ISBN: (纸本)9798331523930

The current paper proposes a new approach for peripheral speech emotion analysis and gender estimation incorporating the best machine learning architectures such as CNNs and LSTMs. Its correct depiction of emotions and the variation of the gendered speech to indicate manifold signal-to-noise ratio make the approach accommodate the difficulty inherent in recognizing emotions from the recording of voices as well as predicting the gender of a speaker. It is then able to ally with preprocessing and collection of a vast range of different speech samples, feature extraction of pitch, tone and spectral data. These features serve as input to the CNNs and LSTMs: LSTMs work well at temporal regular features and the CNNs work well in feature extraction by location (spatio feature). The approach is tested on the RAVDESS and IEMOCAP datasets and establishes a near perfect match to prior approaches with reference to recognition accuracy of emotion analysis and gender prediction. Thus, applying the identified dual method approach, which could be described by such characteristics as highly accurate data pre-preprocessing, clear indication of the features of interest, a different CNN-LSTM model joint training, it offers a reasonably good platform to gain certain insights into such characteristics of the speech as our experiment demonstrates. This work offers satisfactory configurations on concerns such as human computer relation, customer relation and mental health surveillance to suggest that other applications could be achieved in future.

关键词： Training Emotion recognition Accuracy Computational modeling Speech recognition Speech enhancement Predictive models Feature extraction Data models Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Optimizing symbol visibility through displacement

引用

Applied Mathematics and Computation 2025年 505卷

作者： Gärtner, Bernd Kalani, Vishwas M. Reddy, Meghana Meulemans, Wouter Speckmann, Bettina Stojaković, Miloš Department of Computer Science ETH Zürich Switzerland Department of Computer Science and Engineering I.I.T. Delhi India Department of Mathematics and Computer Science TU Eindhoven Netherlands Department of Mathematics and Informatics Faculty of Sciences University of Novi Sad Serbia

In information visualization, the position of symbols often encodes associated data values. When visualizing data elements with both a numerical and a categorical dimension, positioning in the categorical axis admits some flexibility. This flexibility can be exploited to reduce symbol overlap, and thereby increase legibility. In this paper, we initialize the algorithmic study of optimizing symbol legibility via a limited displacement of the symbols. Specifically, we consider closed unit square symbols that need to be placed at specified y-coordinates. We optimize the drawing order of the symbols as well as their x-displacement, constrained within a rectangular container, to maximize the minimum visible perimeter over all squares. If the container has width and height at most 2, there is a point that stabs all squares. In this case, we prove that a staircase layout is arbitrarily close to optimality and can be computed in O(nlog⁡n) time. If the width is at most 2, there is a vertical line that stabs all squares, and in this case, we design a 2-approximation algorithm (assuming fixed container height) that runs in O(nlog⁡n) time. As it turns out that a minimum visible perimeter of 2 is always achievable with a generic construction, we measure this approximation with respect to the visible perimeter exceeding 2. We show that, despite its simplicity, the algorithm gives asymptotically optimal results for certain instances. © 2025 Elsevier Inc.

关键词： Approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

A Multi-Scale Deep Learning Framework Combining MobileViT-ECA and LSTM for Accurate ECG Analysis

引用

IEEE Access 2025年 13卷 85473-85492页

作者： Ba Mahel, Abduljabbar S. Al-Gaashani, Mehdhar S. A. M. Alkanhel, Reem Ibrahim Hassan, Dina S. M. Muthanna, Mohammed Saleh Ali Muthanna, Ammar Aziz, Ahmed University of Electronic Science and Technology of China School of Life Science and Technology Chengdu610054 China University of Electronic Science and Technology of China School of Resources and Environment Chengdu610056 China Princess Nourah bint Abdulrahman University College of Computer and Information Sciences Department of Information Technology P.O. Box 84428 Riyadh11671 Saudi Arabia Tashkent State University of Economics Department of International Business Management Tashkent Uzbekistan RUDN University Department of Applied Probability and Informatics Peoples’Friendship University of Russia 6 Miklukho-Maklaya St Moscow117198 Russia Faculty of Computer and Artificial Intelligence Benha University Department of Computer Science Egypt Central Asian University Engineering school Tashkent Uzbekistan

Electrocardiogram (ECG) analysis is crucial for diagnosing cardiovascular diseases (CVD), especially atrial fibrillation (AF), a prevalent cardiac rhythm abnormality. However, the variability and complexity of ECG signals make AF classification challenging, highlighting the need for more accurate and reliable methods. This paper introduces a novel deep learning (DL) architecture designed to enhance the processing, feature extraction, and analysis of ECG signals. The proposed model utilizes a hybrid framework that combines standard and dilated convolutional networks, advanced attention mechanisms, and temporal sequence learning to address the complexities of ECG data. A parallel architecture integrates a MobileViT block for efficient spatial feature extraction and an Efficient Channel Attention (ECA) module to refine feature representations. Simultaneously, a Long Short-Term Memory (LSTM) network captures temporal dependencies in the data. The concatenation of the outputs from both the MobileViT-ECA block and the LSTM network allows for the extraction of both local and global features. Fully connected layers, along with dropout regularization, ensure robust performance and effective classification. We analyzed ECG signals of three different lengths (1, 2, and 3 seconds) to investigate how segment length affects model performance. Additionally, we conducted various classification scenarios and an ablation study to simulate real-world conditions and identify key components of the architecture. Our experimental findings show that the model attains high accuracy in ECG classification and AF identification, with average metrics including accuracy of 87.80%, recall of 87.80%, F1-score of 87.45%, and specificity of 95.66%. The proposed model outperforms existing methods on the same AF dataset, demonstrating its superior performance. This method accurately classifies ECG signals and detects AF, highlighting its potential for clinical application. The results help move forwar

关键词： Electrocardiography

来源：评论

学校读者我要写书评

暂无评论

A novel real-time noise-resilient zeroing neural network and its applications to matrix problem solving

引用

Mathematics and computers in Simulation 2025年

作者： Yang, Yiguo Wu, Pin Katsikis, Vasilios N. Li, Shuai Feng, Weibing The School of Computer Engineering and Science Shanghai University Shanghai200444 China The Department of Economics Division of Mathematics-Informatics and Statistics-Econometrics National and Kapodistrian University of Athens Athens10559 Greece The Faculty of Information Technology and Electrical Engineering University of Oulu Oulu90570 Finland

Given the critical role of zeroing neural networks (ZNN) in various fields and the practical demand for models in effectively resisting real-time noise, this study introduces a novel anti-noise integral zeroing neural network (AN-IZNN) model alongside its enhanced counterpart (EAN-IZNN), for the applications of matrix problem solving. Theoretical analysis demonstrates their ability to achieve convergence even under different noise conditions. Both theoretical analyses and simulation validations highlight the superior performance of the proposed models over existing neural network models. Notably, the root mean square error of the proposed AN-IZNN and EAN-IZNN models is reduced by 92.6249% and 91.4178%, respectively, compared to scenarios without the proposed method, demonstrating the effectiveness of the solution. © 2025

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

DSLR-CNN: Efficient CNN Acceleration using Digit-Serial Left-to-Right Arithmetic

arXiv

引用

arXiv 2025年

作者： Nisar, Malik Zohaib Ibrahim, Muhammad Sohail Gorgin, Saeid Usman, Muhammad Lee, Jeong-A Department of Computer Engineering College of IT Convergence Chosun University Gwangju61452 Korea Republic of Faculty of Informatics and Data Science University of Regensburg Regensburg93053 Germany Tehran33535-111 Iran

Digit-serial arithmetic has emerged as a viable approach for designing hardware accelerators, reducing interconnections, area utilization, and power consumption. However, conventional methods suffer from performance and latency issues. To address these challenges, we propose an accelerator design using left-to-right (LR) arithmetic, which performs computations in a most-significant digit first (MSDF) manner, enabling digit-level pipelining. This leads to substantial performance improvements and reduced latency. The processing engine is designed for convolutional neural networks (CNNs), which includes low-latency LR multipliers and adders for digit-level parallelism. The proposed DSLR-CNN is implemented in Verilog and synthesized with Synopsys design compiler using GSCL 45nm technology, the DSLR-CNN accelerator was evaluated on AlexNet, VGG-16, and ResNet-18 networks. Results show significant improvements across key performance evaluation metrics, including response time, peak performance, power consumption, operational intensity, area efficiency, and energy efficiency. The peak performance measured in GOPS of the proposed design is 4.37× to 569.11× higher than contemporary designs, and it achieved 3.58× to 44.75× higher peak energy efficiency (TOPS/W), outperforming conventional bit-serial designs. © 2025, CC BY.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Survey on GenAI-Enabled 6G: Technologies, Challenges, and Future Research Avenues

IEEE Open Journal of the Communications Society

引用

IEEE Open Journal of the Communications Society 2025年

作者： Sheraz, Muhammad Chuah, Teong Chee Tareen, Wajahat Ullah khan Al-Habashna, Ala'a Saeed, Sohail Imran Ahmed, Manzoor Lee, It Ee Guizani, Mohsen Multimedia University Faculty of Artificial Intelligence and Engineering Selangor Cyberjaya63100 Malaysia University of Jeddah College of Engineering Department of Electrical and Electronic Engineering Jeddah21589 Saudi Arabia Al Hussein Technical University School of Computing and Informatics Amman Jordan Bahria University Department of Computer Engineering Karachi Pakistan School of Computer Science and Information Engineering Hubei Engineering University Xiaogan432000 China Abu Dhabi United Arab Emirates

The integration of artificial intelligence (AI) in 6G demonstrates a transformative leap in redefining network efficiency, intelligence, and adaptability. However, AI largely leverages discriminative models relying on labelled and quality data, where data accessibility remains serious concern. Generative AI (GenAI) has gained traction due to its immense potential in resolving the issue of data scarcity, complexity, and incompleteness. GenAI models excel in understanding underlying data distributions, enabling them to generate synthetic data that mirrors real-world patterns. GenAI supports adaptive learning and scenario modeling, making it indispensable for addressing the unpredictability and complexity inherent in 6G networks. Since the complexity of wireless communication systems is increasing and the demand for such systems is growing, GenAI presents new ideas for enhancing network performance, increasing system efficiency, and developing intelligent decision-making capabilities. This survey paper investigates the promising role of GenAI in the evolution of 6G networks. An in-depth discussion of notable GenAI models is presented, outlining their application in enhancing key network components. Specifically, the application of GenAI in advanced technologies including reconfigurable intelligent surfaces (RIS), unmanned aerial vehicles (UAVs), digital twins (DTs), and integrated sensing and communications (ISACs) is thoroughly investigated with respect to optimize the adaptability, flexibility, and robustness of the wireless networks. Moreover, use cases of GenAI-enabled wireless networks are presented to highlight the realization of GenAI in 6G. The paper also presents the lessons learned, existing challenges, and future research directions. This paper systematically explores GenAI and its pivotal role in the development of 6G, providing a foundation for researchers to further investigate and advance GenAI-enabled 6G. © 2020 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Adaptive AI-Based Personalized Learning for Accelerated Vocabulary and Syntax Mastery in Young English Learners

引用

International Journal of Advanced computer Science and Applications 2025年第4期16卷 678-687页

作者： Aravind, Angalakuduru Durairaj, M. Chitkara, Preeti Baker El-Ebiary, Yousef A. Muniyandy, Elangovan Ushasree, Linginedi Ammar, Mohamed Ben Kodad Telangana 508206 India Dept.of English Panimalar Engineering College Poonamallee Chennai600123 India Head PR & International Relations Department of Applied Sciences KIET Group of Institutions Delhi-NCR Ghaziabad India Faculty of Informatics and Computing UniSZA University Malaysia Department of Biosciences-Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai602 105 India Applied Science Research Center Applied Science Private University Amman Jordan Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram Andhra Pradesh Guntur Dist. 522302 India Center for Scientific Research and Entrepreneurship Northern Border University Arar73213 Saudi Arabia

Language acquisition is an integral part of early schooling, but young English language learners struggle to learn vocabulary and syntax since they are not provided with specialized instruction. Conventional teaching may vary according to different learning speeds and it leads to unbalanced levels of proficiency among students and possibly leading to disengagement among slow learners. The present computer-assisted learning aids provide practice interactively but without real-time adaptation and personalized feedback, limiting their capacity to address learners’ unique problems. To overcome these constraints, this study suggests an Artificial Intelligence based personalized learning system that supports vocabulary and syntax learning via adaptive learning models, NLP-based chatbots and gamified interactive lessons. The system dynamically adapts content according to students’ most recent performance in real time to enable a personalized learning experience, which results in efficient Learning. The research has experimental study design, and two groups are considered, an AI-supported learning group and a traditional learning group. Pre-test and post-test design measures the effects of the system on vocabulary recall and syntax correctness. Other learner engagement rates like survey results and qualitative feedback inform learner experience and learning efficacy. Initial results indicate that learners working with the Artificial Intelligence powered learning system gained 25percent in recalling vocabulary and 30percent in syntax accuracy over the control group. Further, learner engagement rates are elevated because of real-time feedback and gamification components. These results emphasize the promise of AI-based personalized learning to boost language acquisition and lay the basis for further effective innovations in adaptive education technologies. © (2025), (Science and Information Organization). All rights reserved.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

CrazyKhoreia: A Robotic Perception System for MAV Path Planning From Digital Images

引用

SN computer Science 2025年第5期6卷 1-20页

作者： Restrepo-Garcia, Santiago Romero-Cano, Victor Faculty of Engineering Universidad Autónoma de Occidente Cll 25 115-85 Valle del Cauca Cali 760030 Colombia School of Computer Science and Informatics Cardiff University Wales Cardiff CF24 4 AG United Kingdom

Micro Air Vehicles (MAVs) can be used for a wide range of applications, such as drone-based light shows for cultural or marketing purposes, or as a replacement for fireworks. However, more widespread usage of this technology is precluded by the usability gap between choreography design and deployment on target drones. Seamless deployment of MAV choreographies requires the building of computational interfaces that can obtain drone trajectories from conventional media such as digital images. In this paper, we propose CrazyKhoreia, a low-cost, computationally efficient approach to MAV choreography design. CrazyKhoreia is a robotic perception system that obtains a safe, traversable, and accurate waypoint matrix from a digital image. We validate our trajectory generation system through two distinct modes of operation: light painting, where an MAV flies through all waypoints, and multi-drone formation, in which multiple MAVs are arranged to emulate a given image. The utility of each mode is evaluated differently, using the full resolution CrazyKhoreia output as a reference for comparison. For the light painting mode, we logged the MAV’s pose at a frequency of 10 Hz and compared it with the reference for different levels of detail in the reproduced path. The Root Mean Squared Error (RMSE) ranges from 0.1287 to 0.3190 m. For multi-drone formation, where the swarm remains stationary, we computed the RMSE by comparing observed positions with the reference across multiple tests, resulting in a mean RMSE value of 0.1360 m. © The Author(s) 2025.

关键词： Boundary tracing computer vision Drone swarm MAV Path planning Robotic perception

来源：评论

学校读者我要写书评

暂无评论

Gender Classification from Pashto Handwritten Text Images

引用

IEEE Access 2025年 13卷 81736-81754页

作者： Sultan, Khan Ahmad, Riaz Muhammad, Siraj Almutairi, Sulaiman Abohashrh, Mohammed Ullah, Khalil Namoun, Abdallah Hussain, Ibrar Shaheed Benazir Bhutto University Department of Computer Science Dir Upper Khyber Pakhtunkhwa Sheringal75660 Pakistan Qassim University College of Applied Medical Sciences Department of Health Informatics Qassim52571 Saudi Arabia King Khalid University College of Applied Medical Sciences Department of Basic Medical Sciences Abha61421 Saudi Arabia University of Malakand Department of Software Engineering Dir Lower Khyber Pakhtunkhwa Malakand18800 Pakistan Islamic University of Madinah AI Center Faculty of Computer and Information Systems Madinah42351 Saudi Arabia Shaheed Benazir Bhutto University Sheringal Quality Enhancement Cell Khyber Pakhtunkhwa Dir Upper75660 Pakistan

computer vision (CV) is a subfield of computer science that enables machines to perceive, interpret, and understand visual data. It combines image processing, analysis, and machine learning to extract meaningful insights from images and videos. One key application of CV is the classification of gender from handwritten text images-a complex task, especially in languages such as Pashto due to the intricacy of the script and the limited resources. This research presents a novel approach to gender classification using Pashto Handwritten Text Images (PHTI). The study evaluates and compares the performance of traditional machine learning and modern deep learning models: Support Vector Machine (SVM) and You Only Look Once version 8 (YOLO8), a CNN-based object detection framework. The PHTI dataset, introduced in this study, consists of 36,086 text line images written by 200 male and 200 female native Pashto writers. The goal is to assess how effectively SVM and YOLO8 can classify gender based on visual handwriting patterns in the Pashto language. The experimental results reveal that YOLO8 outperforms SVM in classification accuracy. Although SVM, a widely used machine learning model, achieved an accuracy of 61.29%, YOLO8 reached a significantly higher accuracy of 77% in the same data set. This performance gap highlights the advantages of using advanced deep learning models for complex pattern recognition tasks in CV. This research provides a pioneering contribution by focusing on Pashto, a low-resource language, in the context of gender classification using handwritten text images. It opens new pathways for applying CV in regional languages and supports the development of intelligent systems capable of handling various scripts. © 2013 IEEE.

关键词： Machine vision

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：