检索结果-内蒙古大学图书馆

9th International Conference on Control, Automation, Robotics and Vision

作者： Chaudhari, Narendra S. Ong, Yew Soon Trivedi, Veena Nanyang Technol Univ Sch Comp Engn Singapore 639798 Singapore SGSITS Dept Comp Engn Indore 452003 Madhya Pradesh India

ISBN: (纸本)9781424403417

In theoretical computer science, it is well-established fact that Turing Model (and its equivalent models like Chomsky's phrase structured grammar model, or recursively enumerable functions) is a most general model for computation. Many biologically inspired computing paradigms such as traditional neural networks, genetic algorithms, evolutionary computing frameworks etc., aim at addressing the problem of automatic generation of algorithmic solutions for many computationally difficult problems. Various soft computing models have complementary capabilities. This motivates us to investigate systematic framework for their integration. In this paper, we give an overview of the computational capabilities of some soft computing models, and make some observations about their integration.

关键词： hybrid soft computing neural network optimization genetic algorithms memetic algorithms small world models

来源：评论

学校读者我要写书评

暂无评论

Application of Feed-Forward neural networks for Classifying Acoustics Levels in Vehicle Cabin

Application of Feed-Forward Neural Networks for Classifying ...

引用

4th International Conference on Noise, Vibration and Comfort (NVC 2012)

作者： Junoh, Ahmad Kadri Nopiah, Zulkifli Mohd Ariffin, Ahmad Kamal Univ Malaysia Perlis Inst Engn Math IMRG Kuala Perlis 02000 Perlis Malaysia Univ Kebangsaan Malaysia Fac Engn & Built Environm Dept Mech & Mat Engn Bangi 43600 Selangor Malaysia Univ Kebangsaan Malaysia Ctr Automot Res Fac Engn & Built Environm UKM Bangi 43600 Selangor Malaysia

ISBN: (纸本)9783037859308

Vehicle acoustical comfort and vibration in a passenger car cabin are the main factors that attract a buyer in car purchase. Numerous studies have been carried out by automotive researchers to identify and classify the acoustics level in the vehicle cabin. The objective is to form a special benchmark for acoustics level that may be referred for any acoustics improvement purpose. This study is focused on the sound quality change over the engine speed [rpm] to recognize the noise pattern experienced in the vehicle cabin. Since it is difficult for a passenger to express, and to evaluate the noise experienced or heard in a numerical scale, a neural network optimization approach is used to classify the acoustics levels into groups of noise annoyance levels. A feed forward neural network technique is applied for classification algorithm, where it can be divided into two phases: Learning Phase and Classification Phase. The developed model is able to classify the acoustics level into numerical scales which are meaningful for evaluation purposes.

关键词： Sound Quality Acoustics Level neural network optimization

来源：评论

学校读者我要写书评

暂无评论

Enhanced Grey Wolf optimization for GMPP Tracking of PV Systems Under Partial Shading Condition 21

Enhanced Grey Wolf Optimization for GMPP Tracking of PV Syst...

引用

21st International Middle East Power Systems Conference (MEPCON)

作者： El-sharawy, Latefa A. El-helw, Hadi M. Hasanien, Hany M. Arab Acad Sci Technol & Maritime Transport Elect & Control Dept Cairo Egypt Ain Shams Univ Elect Power & Machine Dept Cairo Egypt

ISBN: (纸本)9781728152899

A maximum power point tracking algorithm (MPPT) for a photovoltaic (PV) system using an Enhanced Grey Wolf optimization (EGWO) technique is presented in this paper. The basic purpose is to study for the mismatches of power in PV modules for non-uniform solar irradiance. MPPT varies the ratio between the voltage and current delivered to the load in order to deliver maximum power. Under partially shaded conditions (PSCs), MPPT algorithm must operate at the Global Maximum Power Point (GMPP). MATLAB/Simulink is used to model and simulate the proposed method. The simulation results reveals that EGWO algorithm outperforms some of the recent algorithms to detect the GMPP under different shading patterns applied to the PV system. EGWO is compared with the hybrid Artificial neural network assisted Perturb and Observe (ANN-P&O) and Multi-Verse optimization (MVO) algorithms. Results show that EGWO outperforms both ANN-P&O and MVO in terms of efficiency and tracking speed.

关键词： Maximum power point tracking Photovoltaic system Partial Shading Conditions neural network optimization Perturb and Observe Maximum Power Point Grey Wolf optimization Artificial Intelligent

来源：评论

学校读者我要写书评

暂无评论

Towards Solving neural networks with optimization Trajectory Search 19

Towards Solving Neural Networks with Optimization Trajectory...

引用

Genetic and Evolutionary Computation Conference (GECCO)

作者： Parsenadze, Lia T. Vargas, Danilo Vasconcellos Fujita, Toshiyuki Kyushu Univ Fukuoka Japan

ISBN: (纸本)9781450367486

Modern gradient based optimization methods for deep neural networks demonstrate outstanding results on image classification tasks. However, methods that do not rely on gradient feedback fail to tackle deep network optimization. In the field of evolutionary computation, applying evolutionary algorithms directly to network weights remains to be an unresolved challenge. In this paper we examine a new framework for the evolution of deep nets. Based on the empirical analysis, we propose the use of linear sub-spaces of problems to search for promising optimization trajectories in parameter space, opposed to weight evolution. We show that linear sub-spaces of loss functions are sufficiently well-behaved to allow trajectory evaluation. Furthermore, we introduce fitness measure to show that it is possible to correctly categorize trajectories according to their distance from the optimal path. As such, this work introduces an alternative approach to evolutionary optimization of deep networks.

关键词： optimization trajectory search search methodologies neural network optimization

来源：评论

学校读者我要写书评

暂无评论

Genetic algorithm for automatic construction, optimization and testing of multi-input forked convolutional neural networks used for diagnostics of rotating agricultural machines 28

Genetic algorithm for automatic construction, optimization a...

引用

28th International Conference on Methods and Models in Automation and Robotics (MMAR)

作者： Sewiolo, Mateusz Mystkowski, Arkadiusz Bialystok Tech Univ Fac Elect Engn PL-15351 Bialystok Poland

ISBN: (纸本)9798350362350;9798350362343

The third decade of the 21st century can undoubtedly be called the "smart" decade. Manufacturers of all devices try to integrate the functions of their products with methods based on artificial intelligence as much as possible. This trend has also reached industrial and agricultural machine manufacturers, where neural network diagnostics is no longer just a thing of the future and is becoming something every day. As developing AI methods requires a high degree of expert knowledge and is time-consuming, we propose a method to automate the entire process. Thanks to the use of a three-step genetic algorithm, we managed to create a system that can create the appropriate AI architecture based on neural networks, optimise it, and test it completely autonomously. We confirmed its effectiveness by using our method to develop a multi-input, forked convolutional network capable of recognizing nine different states (healthy and eight different damages) of an agricultural rotary tedder with an accuracy of approximately 98%.

关键词： convolutional neural network genetic algorithm agriculture rotary tedder vibration analysis fault diagnosis neural network optimization

来源：评论

学校读者我要写书评

暂无评论

Enhanced-AlexNet for Smart Detection of Tomato Seedling Disease

Enhanced-AlexNet for Smart Detection of Tomato Seedling Dise...

引用

2024 IEEE International Conference on Cognitive Computing and Complex Data

作者： Adeshida, Joel Richmond Liu, Lie Abdalla, Ahmed N. Ju, Yongfei Chen, Bo Huaiyin Inst Technol Fac Elect Informat Engn Huaian City Peoples R China

ISBN: (纸本)9798350352900;9798350352894

The accurate identification of seedling diseases in tomato plants is crucial for ensuring optimal crop yields and agricultural productivity. However, traditional image identification methods often rely on visual inspection and experience, can lead to misdiagnosis and inappropriate pesticide use, thereby affecting both yield and quality. This paper addresses this challenges by presenting an Enhanced-AlexNet model designed to improve the identification of tomato seedling diseases through advanced image analysis techniques. The Enhanced-AlexNet model improved the overfitting problem of the standard AlexNet network model by using data augmentation and random inactivation of partial neurons. The model was trained using ten types of tomato data and a total of 4600 images were used, with 70% allocated for training and 30% for testing. Through comprehensive testing, with the original AlexNet model, LeNet model and the ResNet model our proposed model showed better performance than other network models and the overfitting problem of the original AlexNet model was greatly improved after improvement. The accuracy of the Enhanced-AlexNet model is 95.8%, which is 2.4% higher than the original model making it more effective in providing early and accurate disease diagnosis.

关键词： AlexNet Deep learning Tomato leaf diseases neural network optimization Image-based plant disease diagnosis

来源：评论

学校读者我要写书评

暂无评论

Educational Tool-spaces for Convolutional neural network FPGA Design Space Exploration Using High-Level Synthesis 24

Educational Tool-spaces for Convolutional Neural Network FPG...

引用

34th Great Lakes Symposium on VLSI (GLSVLSI)

作者： Yarnell, Richard C. Hossain, Mousam Medina, Raul Graterol Pindoria, Ayush Ghimire, Sujan Chowdhury, Muhtasim Alam Salehi, Soheil Bai, Yu Demara, Ronald F. Univ Cent Florida Dept Elect & Comp Engn Orlando FL 32816 USA Univ Arizona Dept Elect & Comp Engn Tucson AZ USA Calif State Univ Dept Elect & Comp Engn San Bernardino CA USA

ISBN: (纸本)9798400706059

There is significant demand and urgency to prepare electrical and computer engineering students regarding the operational and performance characteristics of machine learning (ML) hardware accelerators. Convolutional neural networks (CNNs), which are utilized for real-time and large dataset image classification tasks, are appropriate targets for hardware acceleration. Designing accelerators for CNNs necessitates understanding the manipulation of CNN parameters. We introduce a hands-on pedagogy whereby learners can identify, modify, and appreciate the interaction of the CNN parameters within an interactive GUI. CASCADE (Computer Aided Student's CNN Analyzer for Design Exploration), a simulation-based framework for Design Space Exploration (DSE) of CNN FPGA-based accelerators is developed, including datapath synthesis, simulation, training, and testbench steps. We offer a case study of High-Level Synthesis (HLS) based CNN implementations targeting the MNIST dataset and present simulation results, namely hardware utilization, accuracy, and operating frequency, and offer insight into potential design trade-offs facing modern engineers.

关键词： High-Level Synthesis neural network optimization Design Space Exploration Educational Frameworks FPGAs STEM curricula

来源：评论

学校读者我要写书评

暂无评论

Advances in Games Technology: Software, Models, and Intelligence

引用

SIMULATION & GAMING 2009年第6期40卷 752-801页

作者： Prakash, Edmond Brindle, Geoff Jones, Kevin Zhou, Suiping Chaudhari, Narendra S. Wong, Kok-Wai Manchester Metropolitan Univ Manchester Lancs England Nanyang Technol Univ Singapore Singapore Murdoch Univ Murdoch WA Australia

Games technology has undergone tremendous development. In this article, the authors report the rapid advancement that has been observed in the way games software is being developed, as well as in the development of games content using game engines. One area that has gained special attention is modeling the game environment such as terrain and buildings. This article presents the continuous level of detail terrain modeling techniques that can help generate and render realistic terrain in real time. Deployment of characters in the environment is increasingly common. This requires strategies to map scalable behavior characteristics for characters as well. The authors present two important aspects of crowd simulation: the realism of the crowd behavior and the computational overhead involved. A good simulation of crowd behavior requires delicate balance between these aspects. The focus in this article is on human behavior representation for crowd simulation. To enhance the player experience, the authors present the concept of player adaptive entertainment computing, which provides a personalized experience for each individual when interacting with the game. The current state of game development involves using very small percentage (typically 4% to 12%) of CPU time for game artificial intelligence (AI). Future game AI requires developing computational strategies that have little involvement of CPU for online play, while using CPU's idle capacity when the game is not being played, thereby emphasizing the construction of complex game AI models offline. A framework of such nonconventional game AI models is introduced.

关键词： artificial intelligence behavior modeling board game character modeling conversational avatars environment modeling game AI game development game editor game engine game software gardening game genetic algorithms hybrid soft computing memetic algorithms neural network optimization programmable tournament racing game small world models terrain modeling

来源：评论

学校读者我要写书评

暂无评论

PO-DARTS: Post-optimizing the Architectures Searched by Differentiable Architecture Search Algorithms 1

引用

32nd International Conference on Artificial neural networks (ICANN)

作者： Hao, Debei Pei, Songwei Beijing Univ Posts & Telecommun Natl Pilot Software Engn Sch Sch Comp Sci Beijing 100876 Peoples R China

ISBN: (数字)9783031442018

ISBN: (纸本)9783031442001;9783031442018

In neural architecture search, differentiable architecture search algorithm has become one of the mainstream methods. However, no matter in the search or evaluation stage, the architecture is repeatedly stacked by two kinds of Cells, namely Normal Cell and Reduction Cell, respectively. This undoubtedly limits the performance of the evaluation architecture to a large extent due to the architecture restriction, resulting in sub-optimal performance. In order to alleviate the impact of architecture restriction on network performance, this paper proposes to post-optimize the architecture searched by differentiable architecture search algorithms by freezing the architecture parameters of partial Cells and further searching other Cells to bring more diversity into the stacked Cells. The proposed post-optimizing methods consist of the global post-optimizing search method and the local post-optimizing search method, respectively. The performance of the evaluation architecture can benefit from the diverse stacked Cells with less architecture restriction. In the experiments, the proposed post-optimizing method is applied to the mainstream differentiable architecture search algorithms such as DARTS and P-DARTS, and superior results have been achieved on CIFAR-10 and CIFAR-100 datasets. Moreover, the proposed method can obtain the post-optimized architecture with limited computing resources.

关键词： Deep Learning Image Classification Differentiable Architecture Search neural network optimization

来源：评论

学校读者我要写书评

暂无评论

RD-Crack: A Study of Concrete Crack Detection Guided by a Residual neural network Improved Based on Diffusion Modeling 33rd

RD-Crack: A Study of Concrete Crack Detection Guided by a Re...

引用

33rd International Conference on Artificial neural networks and Machine Learning (ICANN)

作者： Huang, Yubo Lai, Xin Wang, Zixi Ye, Muyang Li, Yinmian Li, Yi Zhang, Fang Luo, Chenyang Guangxi Univ Sch Comp Elect & Informat Nanning 530004 Peoples R China Southwest Jiaotong Univ Sch Civil Engn Chengdu 611756 Sichuan Peoples R China Southwest Jiaotong Univ Sch Comp & Artificial Intelligence Chengdu 611756 Sichuan Peoples R China Southwest Jiaotong Univ SWJTU Leeds Joint Sch Chengdu 611756 Sichuan Peoples R China Univ Lancaster Comp & Commun Lancaster England

ISBN: (纸本)9783031723551;9783031723568

Automated crack detection in concrete structures is an important aspect of structural health monitoring (SHM) to ensure safety and durability. Traditional methods mainly rely on manual inspection, which suffers from subjectivity and inefficiency challenges. To address these issues, machine learning, especially deep learning techniques, has been gradually adopted to improve accuracy and reduce reliance on large amounts of labeled data. This paper introduces RD-Crack, an innovative concrete crack detection method. Our RD-Crack framework combines the encoder with ResNeXt and extrusion excitation modules for feature extraction and uses a diffusion model for parameter optimization to achieve accurate crack detection in complex engineering environments. Experimental results show that our RD-Crack outperforms other state-of-the-art methods in comprehensive performance.

关键词： Crack Detection Structural Health Monitoring Diffusion Model Unsupervised Learning neural network optimization

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：