检索结果-内蒙古大学图书馆

Conference on High-Performance Computing in Geoscience and Remote Sensing VI

作者： Li, Ligang Ni, Wei Ma, Xiaoshan Yang, Zhen Meng, Xin Shen, Feifei Chinese Acad Sci Natl Space Sci Ctr Key Lab Elect & Informat Technol Space Syst Beijing 100190 Peoples R China

ISBN: (纸本)9781510604186;9781510604193

The simulation of high accuracy three-dimension (3D) scene optical field radiation distribution can provide the input for camera design, optimization of key parameters and testing of various imaging models. It can benefit for reducing the strong coupling between the imaging models and scene simulation. However, the simulation computation is extremely large and the non-optimization computing method can't performed efficiently. Therefore, a study was carried out from the algorithm optimization and using high-performance platform to accelerate the operation speed. On the one hand, the visibility of scene was pre-computed which include the visibility from the light source to each facet in scene and the visibility between facets. The bounding box accelerate algorithm was adopted which can avoid a lot of time-consuming computation of occlusion in the light field radiation simulation process. On the other hand, since the 3D scene light field radiation simulation was obtained by a large number of light approximation, the algorithms can be divided blocks and processed parallelly. The GPU parallel framework was adopted to realize the simulation model of light field radiation in visible band. Finally, experiments were performed. The result shown the proposed method was more efficient and effective compared with the non-optimization method.

关键词： computation optimization visible light band light field radiation simulation three-dimension scene

来源：评论

学校读者我要写书评

暂无评论

A 3D virtual geographic environment for flood representation towards risk communication

引用

INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION 2024年 128卷

作者： Li, Weilian Zhu, Jun Pirasteh, Saied Zhu, Qing Guo, Yukun Luo, Lan Dehbi, Youness Southwest Jiaotong Univ Fac Geosiences & Engn Chengdu 611756 Peoples R China HafenCity Univ Hamburg Computat Methods Lab D-20457 Hamburg Germany Shaoxing Univ Inst Artificial Intelligence Shaoxing 312000 Peoples R China Guangdong Hong Kong Macau Joint Lab Smart Cities Shenzhen 518060 Peoples R China

Risk communication seeks to develop a shared understanding of disaster among stakeholders, thereby amplifying public awareness and empowering them to respond more effectively to emergencies. However, existing studies have overemphasized specialized numerical modelling, making the professional output challenging to understand and use by non -research stakeholders. In this context, this article proposes a 3D virtual geographic environment for flood representation towards risk communication, which integrates flood modelling, parallel computation, and 3D representation in a pipeline. Finally, a section of the Rhine River in Bonn, Germany, is selected for experiment analysis. The experimental results show that the proposed approach is capable of flood modelling and 3D representation within a few hours, the parallel speedup ratio reached 6.45. The intuitive flood scene with 3D city models is beneficial for promoting flood risk communication and is particularly helpful for participants without direct experience of floods to understand its spatiotemporal process. It also can be embedded in the Geospatial Infrastructure Management Ecosystem (GeoIME) cloud application for intelligent flood systems.

关键词： Floods Spatiotemporal modelling computation optimization 3D representation Risk communication Virtual geographic environments

来源：评论

学校读者我要写书评

暂无评论

Image Neural Network Classifier by Detail Attributes

引用

PATTERN RECOGNITION AND IMAGE ANALYSIS 2023年第2期33卷 179-183页

作者： Sai, S. V. Nikonov, V. S. Pacific Natl Univ Khabarovsk 680035 Russia

The paper presents the features of the development of a neural network analyzer of image detail. The neural network analyzer is a binary classifier based on the algorithm of search and identification of small image structures with different levels of detail (FDL). The developed neural network has a small number of neurons, requires a small amount of data and can be applied in embedded devices. The neural network classifier is considered in the context of neural network applications for computation optimization. Examples of using the classifier in image analysis and digital processing tasks are given.

关键词： neural network image analysis machine learning computation optimization

来源：评论

学校读者我要写书评

暂无评论

Design of Multi-Class Optimized Lightweight Convolution Neural Network for Rice Classification 7

Design of Multi-Class Optimized Lightweight Convolution Neur...

引用

7th International Women in Data Science Conference at Prince Sultan University (WiDS-PSU)

作者： Deepika, S. Arunachalam, V Vellore Inst Technol Dept Micro & Nanoelect Vellore Tamil Nadu India

ISBN: (纸本)9798350395839;9798350395846

Worldwide, more than 40k rice varieties exist, each with different nutritional content and quality. Identifying these has to be consistent, automated, and accurate. Considering the feature extraction process, convolution Neural Networks (CNN) are preferred over machine learning (ML) for this classification. Transfer learning approaches help to optimize the CNN model;therefore, it fits in an FPGA. Seven different CNN models were proposed to classify five rice varieties, each model differs based on the: kernel depth;the number of convolution layers (CL);the number of fully connected layers (FCL);and the number of neurons per FCL. These were analyzed considering 70% and 30% for training and testing respectively. A dataset of 15,000 images/variety with each image of resolution. This results in an Optimized Lightweight Convolutional Neural Network (OpLW-CNN) model, having a CL, followed by two FCLs. This model is further analyzed using a random set of images: 500, 5000, and 75000 to fit the model optimally. This model achieves 99%, 98.13%, and 98.14% specificity, F1-score, and accuracy for a set of 5000 images. These metrics are approximately 1% to 2% lesser than the performance of the benchmark model, and 81.5% fewer computations. Also, this model requires less than a second to classify an image.

关键词： Convolution Neural Network computation optimization Multi-class Classifier Rice Classification Transfer Learning

来源：评论

学校读者我要写书评

暂无评论

optimization and Deployment of Memory-Intensive Operations in Deep Learning Model on Edge

引用

计算机科学 2023年第2期50卷 3-12页

作者： Peng XU Jianxin ZHAO Chi Harold LIU Department of Computer Science and Technology Beijing Institute of TechnologyBeijing 100081China

As a large amount of data is increasingly generated from edge devices,such as smart homes,mobile phones,and wearable devices,it becomes crucial for many applications to deploy machine learning modes across edge *** execution speed of the deployed model is a key element to ensure service *** a highly heterogeneous edge deployment scenario,deep learning compiling is a novel approach that aims to solve this *** defines models using certain DSLs and generates efficient code implementations on different hardware ***,there are still two aspects that are not yet thoroughly investigated *** first is the optimization of memory-intensive operations,and the second problem is the heterogeneity of the deployment *** that end,in this work,we propose a system solution that optimizes memory-intensive operation,optimizes the subgraph distribution,and enables the compiling and deployment of DNN models on multiple *** evaluation results show the performance of our proposed system.

关键词： Memory optimization Deep compiler computation optimization Model deployment Edge computing

来源：评论

学校读者我要写书评

暂无评论

Optimizing Transformer Training Based on computation and Accessing Memory Features with Deep Learning Processor 29

Optimizing Transformer Training Based on Computation and Acc...

引用

29th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2023

作者： Lei, Zhou Fang, Qin Qian, Jingfeng Chen, Shengbo Xu, Qingguo Yang, Ninghua Shanghai University School of Computer Engineering and Science Shanghai200444 China Ningbo Zhongke Intelligent Computing Technology Co. Ltd Zhejiang315100 China

ISBN: (纸本)9798350330717

The Transformer model, which has significantly advanced natural language processing and computer vision, overcomes the limitations of recurrent neural networks and convolutional neural networks. However, it faces challenges with computational efficiency and memory management due to complex computations and variable-length inputs. Despite research efforts, these issues persist. This paper presents a novel optimization of the Transformer model, following an indepth analysis of its computational graph structure. Firstly, we utilize Deep Computing Unit (DCU) as our hardware platform. Secondly, we optimize element-wise and reduction operators through operator fusion and rewriting. Thirdly, we develop a fine-grained memory management algorithm using a greedy strategy. As a result, the training speed of the Transformer model increases by 1.2x - 1.4x without compromising accuracy. © 2023 IEEE.

关键词： computation optimization DCU memory management Transfomer

来源：评论

学校读者我要写书评

暂无评论

An ultra-low-computation model for understanding sign languages

引用

EXPERT SYSTEMS WITH APPLICATIONS 2024年第PartC期249卷

作者： Fallah, Mohammad K. Najafi, Mohammadreza Gorgin, Saeid Lee, Jeong-A. Chosun Univ Dept Comp Engn Gwangju South Korea

In artificial intelligence applications, advanced computational models, such as deep learning, are employed to achieve high accuracy, often requiring the execution of numerous operations. Conversely, lightweight computational models are typically more resource -efficient, making them suitable for various devices, including smartphones, tablets, and wearable technology. This paper presents an ultra -low -computation solution for interpreting sign languages to assist deaf and hard -of -hearing individuals without needing specialized hardware or significant computational resources. The proposed approach initially performs data abstraction on the input data. During this process, the image is systematically scanned from various perspectives, and the collected information is then encoded into a one-dimensional vector. Subsequently, the abstracted information undergoes processing through a Fully Connected Neural Network (FCN), resulting in highly accurate output. We also introduced two abstraction methods, namely Opaque and Glass, inspired by the interaction of light with different types of objects. The proposed abstractions facilitate the comprehension of the hand gesture's outer boundary as well as its row -wise and column -wise density of pixels. Our experiments on three datasets confirm the efficiency of the proposed method, achieving an accuracy of 99.4% in recognizing American Sign Language, 99.96% accuracy in recognizing Indian Sign Language, and 99.95% accuracy in recognizing Bangla Sign Language. Notably, the model size and the number of MAC operations are significantly smaller than state-of-the-art computational models trained on the same datasets.

关键词： Sign language detection Hand gesture classification computation optimization Data abstraction

来源：评论

学校读者我要写书评

暂无评论

On the minimal wind directions required to assess mean annual air pollution concentration based on CFD results

引用

SUSTAINABLE CITIES AND SOCIETY 2021年 71卷 102920-102920页

作者： Jurado, Xavier Reiminger, Nicolas Vazquez, Jose Wemmert, Cedric AIR&D F-67000 Strasbourg France Univ Strasbourg CNRS ICUBE Lab F-67000 Strasbourg France

computational fluid dynamics has shown a great interest among the scientific community to assess air pollutant concentrations in urban areas and, define strategies to limit air pollution and achieve sustainable cities of the future. Recent studies have given methodologies on how to assess mean annual concentrations based on numerical model results to compare with the annual air quality standards. Nonetheless, these methodologies need many wind directions to be modelled and, therefore, lead to high calculation costs. The purpose of this paper is to present two approaches to decrease the calculation costs when calculating annual concentration from computational fluid dynamics results by (1) ignoring uniformly spaced wind direction and (2) considering the predominant wind directions. According to the results, the first approach is on overall better than the second one for any wind rose or building layout considered. With the first approach, the calculation costs can be reduced up to 50 % without leading to more than 20 % of error, and even less error can be expected for homogeneous wind roses. Finally, a method to finely evaluate errors made when using the first approach versus using the whole wind rose, without computing it, is presented.

关键词： CFD Air pollution Mean annual concentration Wind rose computation optimization

来源：评论

学校读者我要写书评

暂无评论

Extremely Low-bit Convolution optimization for Quantized Neural Network on Modern Computer Architectures 20

Extremely Low-bit Convolution Optimization for Quantized Neu...

引用

49th International Conference on Parallel Processing (ICPP)

作者： Han, Qingchang Hu, Yongmin Yu, Fengwei Yang, Hailong Liu, Bing Hu, Peng Gong, Ruihao Wang, Yanfei Wang, Rui Luan, Zhongzhi Qian, Depei Beihang Univ Beijing Peoples R China SenseTime Res Beijing Peoples R China

ISBN: (纸本)9781450388160

With the continuous demand for higher accuracy of deep neural networks, the model size has increased significantly. Quantization is one of the most widely used model compression methods, which can effectively reduce the model size without severe accuracy loss. Modern processors such as ARM CPU and NVIDIA GPU have already provided the support of low-bit arithmetic instructions. However, there lack efficient and practical optimizations for convolution computation towards extremely low-bit on ARM CPU (e.g., 2 similar to 8-bit) and NVIDIA GPU (e.g., 4-bit and 8-bit). This paper explores the performance optimization methods of extremely low-bit convolution on diverse architectures. On ARM CPU, we propose two instruction schemes for 2 similar to 3-bit and 4 similar to 8-bit convolution with corresponding register allocation methods. In addition, we re-design the GEMM computation with data padding and packing optimizations. We also implement winograd algorithm for convolution with some specific bit width (e.g., 4 similar to 6-bit) to achieve higher performance. On NVIDIA GPU, we propose a data partition mechanism and multi-level memory access optimizations, to better adapt the computation to GPU thread and memory hierarchy. We also propose quantization fusion to eliminate unnecessary data access. The experiment results demonstrate our implementations achieve better performance of extremely low-bit convolution compared to the state-of-the-art frameworks and libraries such as ncnn and cuDNN. To the best of our knowledge, this is the first work that provides efficient implementations of extremely low-bit convolutions covering 2 similar to 8-bit on ARM CPU and 4-bit/8-bit on NVIDIA GPU.

关键词： Extremely Low-bit Convolution Quantized Neural Network computation optimization ARM CPU NVIDIA GPU

来源：评论

学校读者我要写书评

暂无评论

Implementation of Exploration in TONIC Using Non-stationary Volatile Multi-arm Bandits 8th

Implementation of Exploration in TONIC Using Non-stationary ...

引用

8th International Conference on Soft Computing for Problem Solving (SocProS)

作者： Shaha, Aditya Arya, Dhruv Tripathy, B. K. VIT Vellore Sch Comp Sci & Engn Vellore Tamil Nadu India VIT Vellore Sch Informat Technol & Engn Vellore Tamil Nadu India

ISBN: (纸本)9789811500350;9789811500343

Target Oriented Network Intelligence Collection (TONIC) is a problem which deals with acquiring maximum number of profiles in the online social network so as to maximize the information about a given target through these profiles. The acquired profiles, also known as leads in this paper, are expected to contain information which is relevant to the target *** the past, TONIC problem has been solved by modelling it as a Volatile Multi-arm bandit problem with stationary reward distribution. The problem with this approach is that the underlying reward distribution in case of TONIC changes with each exploration which needs to be incorporated for making future acquisitions. The paper shows a successful solution to the TONIC problem by modelling it as Volatile Bandit problem with non-stationary reward distributions. It illustrates a new approach and compares it's performance with other algorithms.

关键词： Volatile multi-arm bandit problems Exploration-exploitation dilemma Discounted upper confidence bound computation optimization Reinforcement 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：