检索结果-内蒙古大学图书馆

42nd IEEE/ACM International Conference on Computer-Aided Design (ICCAD)

作者： Luo, Erjing Huang, Haitong Liu, Cheng Li, Guoyu Yang, Bing Wang, Ying Li, Huawei Li, Xiaowei Chinese Acad Sci Inst Comp Technol SKLP Beijing Peoples R China Beijing Inst Technol Sch Informat & Elect Beijing Peoples R China Univ Chinese Acad Sci Dept Comp Sci Beijing Peoples R China Harbin Univ Sci Technol Dept Comp Sci & Technol Harbin Peoples R China

ISBN: (纸本)9798350322255

Mixed-precision neural networks (MPNNs) that enable the use of just enough data width for a deep learning task promise significant advantages of both inference accuracy and computing overhead. FPGAs with fine-grained reconfiguration capability can adapt the processing with distinct data width and models, and hence, can theoretically unleash the potential of MPNNs. Nevertheless, commodity DPUs on FPGAs mostly emphasize generality and have limited support for MPNNs especially the ones with lower data width. In addition, primitive DSPs in FPGAs usually have much larger data width than that is required by MPNNs and haven't been sufficiently co-explored with MPNNs yet. To this end, we propose an open source MPNN accelerator design framework specifically tailored for FPGAs. In this framework, we have a systematic DSP-packing algorithm to pack multiple lower data width MACs in a single primitive DSP and enable efficient implementation of MPNNs. Meanwhile, we take DSP packing efficiency into consideration with MPNN quantization within a unified neural network architecture search (NAS) framework such that it can be aware of the DSP overhead during quantization and optimize the MPNN performance and accuracy concurrently. Finally, we have the optimized MPNN fine-tuned to a fully pipelined neural network accelerator template based on HLS and make best use of available resources for higher performance. Our experiments reveal the resulting accelerators produced by the proposed framework can achieve overwhelming advantages in terms of performance, resource utilization, and inference accuracy for MPNNs when compared with both handcrafted counterparts and prior hardware-aware neural network accelerators on FPGAs.

关键词： DSP packing mixed precision neural network neural network architecture search quantization and implementation co-optimization

来源：评论

学校读者我要写书评

暂无评论

Automatic neural network Construction-Based Channel Estimation for IRS-Aided Communication Systems

Automatic Neural Network Construction-Based Channel Estimati...

引用

IEEE Wireless Communications and networking Conference (WCNC)

作者： Shi, Haoqing Ji, Taotao Zhang, Zhengming Yang, Luxi Huang, Yongming Southeast Univ Natl Mobile Commun Res Lab Nanjing 210096 Peoples R China Purple Mt Labs Pervas Commun Res Ctr Nanjing 211111 Peoples R China

ISBN: (纸本)9781665491228

Accurate channel estimation is an indispensable prerequisite for intelligent reflecting surface (IRS) aided communication systems to achieve huge system performance gains. Current works show that deep neural network-based channel estimation is a promising solution to achieve competitive performance compared with the conventional methods. However, neural network-based approaches generally realize the channel estimation by manually designing network architectures in a trial-and-error manner which need complex neural network domain knowledge and tremendous computation resource. This paper automatically constructs a high-performance neural network architecture to obtain dedicated channel estimation schemes intelligently. Specifically, we propose a channel estimation neural network architecture search (CENAS) method based on gradient alternatively search strategy to search a channel estimation neural network. With the search space designed meticulously for the channel estimation task, the network searched by the proposed method outperforms the conventional and deep learning-based channel estimation algorithms.

关键词： Intelligent reflecting surface channel estimation neural network architecture search deep learning

来源：评论

学校读者我要写书评

暂无评论

Automatic neural network Design of Scene-customization for Massive MIMO CSI Feedback 98

Automatic Neural Network Design of Scene-customization for M...

引用

98th IEEE Vehicular Technology Conference (VTC-Fall)

作者： Li, Xiangyi Guo, Jiajia Wen, Chao Kai Tian, Wenqiang Jin, Shi Southeast Univ Natl Mobile Commun Res Lab Nanjing 210096 Peoples R China Natl Sun Yat Sen Univ Inst Commun Engn Kaohsiung 80424 Taiwan OPPO Dept Stand Res Beijing 100026 Peoples R China

ISBN: (纸本)9798350329285

Deep learning has revolutionized the design of channel state information (CSI) feedback modules in wireless communication. However, designing an optimal neural network (NN) architecture for CSI feedback can be laborious and time-consuming, especially for customized networks targeting different scenarios. To address this challenge, this paper proposes the use of neural architecture search (NAS) to automatically generate scenario-specific CSI feedback neural network architectures. By employing automated machine learning and gradient-based NAS, an efficient and cost-effective architecture design process is achieved with reduced reliance on expert knowledge and design time, thus lowering the design threshold. This approach leverages implicit scenario knowledge and integrates it into the scenario customization process in a data-driven manner, fully harnessing the potential of deep learning in a given scenario. Experimental results demonstrate that the generated architecture called Auto-CsiNet outperforms manually designed models in terms of reconstruction performance (improvement by approximately 14%) and complexity reduction (approximately 50%), highlighting the effectiveness of NAS-based automated solutions.

关键词： Massive MIMO CSI feedback deep learning neural network architecture search

来源：评论

学校读者我要写书评

暂无评论

A neural network architecture optimizer based on DARTS and generative adversarial learning

引用

INFORMATION SCIENCES 2021年 581卷 448-468页

作者： Zhang, Ting Waqas, Muhammad Shen, Hao Liu, Zhaoying Zhang, Xiangyu Li, Yujian Halim, Zahid Chen, Sheng Beijing Univ Technol Fac Informat Technol Engn Res Ctr Intelligent Percept & Autonomous Con Beijing 100124 Peoples R China GIK Inst Engn Sci & Technol Fac Comp Sci & Engn Topi 23460 Pakistan Res Inst China Telecom Corp LTD Beijing 100032 Peoples R China Univ Southampton Sch Elect & Comp Sci Southampton SO17 1BJ Hants England King Abdulaziz Univ Jeddah 21589 Saudi Arabia

neural network architecture search automatically configures a set of network architectures according to the targeted rules. Thus, it relieves the human-dependent effort and repetitive resources consumption for designing neural network architectures and makes the task of finding the optimum network architecture with better performance much more accessible. network architecture search methods based on differentiable architecture search (DARTS), however, introduces parameter redundancy. To address this issue, this work presents a novel method for optimizing network architectures that combines DARTS with generative adversarial learning (GAL). We first find the module structures utilizing the DARTS algorithm. Afterwards, the retrieved modules are stacked to derive the initial neural network architecture. Next, the GAL is used to prune some branches of the initial neural network, thereby obtaining the final neural network architecture. The proposed DARTS-GAL method re-optimizes the network architecture searched by DARTS to simplify the network connection and reduce network parameters without compromising network performance. Experimental results on benchmark datasets, i.e., Mixed National Institute of Standards and Technology (MNIST), FashionMNIST, Canadian Institute for Advanced Research10 (CIFAR10), Canadian Institute for Advanced Research100 (CIAFR100), Cats vs Dogs, and voiceprint recognition datasets, indicate that the test accuracies of the DARTS-GAL are higher than those of the DARTS in the majority of the cases. In particular, the proposed solution exhibits an improvement in accuracy by 7.35% on CIFAR10 compared with DARTS, attaining the state-of-the-art result of 99.60%. Additionally, the number of network parameters derived by the DARTS-GAL is significantly lower than that by the DARTS method, with a pruning rate of 62.3% at the highest case. (c) 2021 Elsevier Inc. All rights reserved.

关键词： Deep learning neural network architecture search Differentiable architecture search (DARTS) Pruning Generative adversarial learning (GAL)

来源：评论

学校读者我要写书评

暂无评论

Estimation of spatial uncertainty in material property distributions within heterogeneous structures using optimized convolutional neural networks

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2023年第PartB期117卷

作者： Pitz, Emil Rooney, Sean Pochiraju, Kishore Stevens Inst Technol Dept Mech Engn Hoboken NJ 07030 USA

Variations in material behaviors within structures built with heterogeneous materials lead to damage initiation and evolution in locally weak regions. Quantifying the property variability within the structure and forward propagation of the impact of the material property uncertainty on the structural response is critical for reliability analysis and structural performance maximization. Commonly, quantification of the variability requires either computationally expensive high-fidelity models of the underlying microstructure or extensive experimental testing. In this paper, we model the uncertainty with spatially correlated random fields and calibrate the model parameters from limited strain field observations using neural networks (NNs). The calibration is performed by trained NNs, which outputs best-fit parameters for the spatial correlation model by accepting filtered Digital Image Correlation (DIC) strain distributions as the input. We demonstrate that by training the NNs using simulated data, the resulting networks are able to calibrate the spatial distribution uncertainty models effectively for a set of Fused Filament Fabrication (FFF) printed structures. The methodology requires a limited number of experimental datasets and produces fast estimations of the best-fit parameters of the uncertainty model compared to optimization or inverse fitting methods. This method allows experimentalists to use the same DIC information routinely obtained during modulus or strength testing to calibrate a spatial property distribution uncertainty model for the underlying microstructure.

关键词： Uncertainty quantification neural network architecture search Digital image correlation Karhunen-Lo?ve expansion AI-based calibration

来源：评论

学校读者我要写书评

暂无评论

An automatic selection of optimal recurrent neural network architecture for processes dynamics modelling purposes

引用

APPLIED SOFT COMPUTING 2022年第0期116卷 108375-108375页

作者： Laddach, Krzysztof Langowski, Rafal Rutkowski, Tomasz A. Puchalski, Bartosz Gdask Univ Technol Dept Intelligent Control & Decis Support Syst G Narutowicza 11-12 PL-80233 Gdask Poland

A problem related to the development of algorithms designed to find the structure of artificial neural network used for behavioural (black-box) modelling of selected dynamic processes has been addressed in this paper. The research has included four original proposals of algorithms dedicated to neural network architecture search. Algorithms have been based on well-known optimisation techniques such as evolutionary algorithms and gradient descent methods. In the presented research an artificial neural network of recurrent type has been used, whose architecture has been selected in an optimised way based on the above-mentioned algorithms. The optimality has been understood as achieving a trade-off between the size of the neural network and its accuracy in capturing the response of the mathematical model under which it has been learnt. During the optimisation, original specialised evolutionary operators have been proposed. The research involved an extended validation study based on data generated from a mathematical model of the fast processes occurring in a pressurised water nuclear reactor. (C) 2021 Elsevier B.V. All rights reserved.

关键词： Black-box model Evolutionary algorithm neural modelling neural network architecture search Pressurised water reactor

来源：评论

学校读者我要写书评

暂无评论

High-Performance VOC Quantification for IAQ Monitoring Using Advanced Sensor Systems and Deep Learning

引用

ATMOSPHERE 2021年第11期12卷 1487页

作者： Robin, Yannick Amann, Johannes Baur, Tobias Goodarzi, Payman Schultealbert, Caroline Schneider, Tizian Schuetze, Andreas Saarland Univ Lab Measurement Technol Campus A5 1 D-66123 Saarbrucken Germany

With air quality being one target in the sustainable development goals set by the United Nations, accurate monitoring also of indoor air quality is more important than ever. Chemiresistive gas sensors are an inexpensive and promising solution for the monitoring of volatile organic compounds, which are of high concern indoors. To fully exploit the potential of these sensors, advanced operating modes, calibration, and data evaluation methods are required. This contribution outlines a systematic approach based on dynamic operation (temperature-cycled operation), randomized calibration (Latin hypercube sampling), and the use of advances in deep neural networks originally developed for natural language processing and computer vision, applying this approach to volatile organic compound measurements for indoor air quality monitoring. This paper discusses the pros and cons of deep neural networks for volatile organic compound monitoring in a laboratory environment by comparing the quantification accuracy of state-of-the-art data evaluation methods with a 10-layer deep convolutional neural network (TCOCNN). The overall performance of both methods was compared for complex gas mixtures with several volatile organic compounds, as well as interfering gases and changing ambient humidity in a comprehensive lab evaluation. Furthermore, both were tested under realistic conditions in the field with additional release tests of volatile organic compounds. The results obtained during field testing were compared with analytical measurements, namely the gold standard gas chromatography mass spectrometry analysis based on Tenax sampling, as well as two mobile systems, a gas chromatograph with photo-ionization detection for volatile organic compound monitoring and a gas chromatograph with a reducing compound photometer for the monitoring of hydrogen. The results showed that the TCOCNN outperforms state-of-the-art data evaluation methods, for example for critical pollutants such as formaldeh

关键词： volatile organic compounds (VOCs) indoor air quality (IAQ) deep neural networks neural network architecture search temperature-cycled operation (TCO)

来源：评论

学校读者我要写书评

暂无评论

A view on automated neural graph topology generation and a viable direction of innovation 19

A view on automated neural graph topology generation and a v...

引用

19th RoEduNet Annual Conference on networking in Education and Research (RoEduNet)

作者： Damian, Andrei Ionut Piciu, Laurentiu Tapus, Nicolae Lummetry AI Bucharest Romania Univ Politehn Bucuresti Bucharest Romania

ISBN: (纸本)9780738112657

Training deep neural networks requires knowledge and ample experimental time, as well as computational resources due to the search process of the architectural design. In this work, we review the current main research directions in the area of network architecture and finally propose in contrast a novel architecture, namely MultiGatedUnit, that uses directly learnable self-gating mechanisms for automated graph topology generation, currently in research and experimentation phase.

关键词： neural network architecture search deep neural graph topology design

来源：评论

学校读者我要写书评

暂无评论

A Novel Encoding Scheme for Complex neural architecture search 34

A Novel Encoding Scheme for Complex Neural Architecture Sear...

引用

34th International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC)

作者： Ahmad, Mobeen Abdullah, Muhammad Han, Dongil Sejong Univ Dept Comp Engn Vis & Image Proc Lab Seoul South Korea

ISBN: (纸本)9781728132716

Recent boom in Image Processing can be attributed mostly to Artificial neural networks, enormity of data and availability of high-performance hardware. ANNs need to be designed with care and a lot of tweaking is required according to the problem at hand. However, they can be designed automatically by deploying search algorithms given a set of possible values for hyperparameters. Previously, network design has been limited because of the inherent complexity of network diversity in terms of their style (e.g. VGG-style or ResNet-style). Aforementioned two network styles require different kinds of parameters hence limiting the search algorithm to output only a specific style of architecture. We propose a methodology which will allow the system to search for the best architecture with least possible constraints.

关键词： neural network architecture search Functional Blocks Encoding schemes

来源：评论

学校读者我要写书评

暂无评论

Training time reduction for network architecture search using genetic algorithm 14

Training time reduction for network architecture search usin...

引用

14th International Conference on Quality Control by Artificial Vision

作者： Matsuoka, Yuki Aizawa, Hiroaki Sato, Junya Kato, Kunihito Gifu Univ 1-1 Yanagido Gifu Gifu Japan

ISBN: (数字)9781510630543

ISBN: (纸本)9781510630543

Designing the optimal architecture of neural networks is an important issue. However, since this is difficult even for experienced experts, automatic optimization of the network architecture is required. In this study, we regard this issue as a combinatorial optimization problem, and utilize genetic algorithm to optimize the network architecture. Because training the networks, which are represented by individuals in GA, takes a long time, a novel method to reduce the training time by inheriting the weights of the trained network is proposed. From experimental results, our proposed method achieved the time reduction and higher accuracy than a conventional method.

关键词： Genetic Algorithm neural network architecture search convolutional neural network

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：