检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

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

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

16,501 篇 会议
212 篇 期刊文献

馆藏范围

16,713 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

10,815 篇 工学
- 7,694 篇 计算机科学与技术...
- 4,139 篇 电气工程
- 3,533 篇 软件工程
- 3,156 篇 信息与通信工程
- 1,021 篇 控制科学与工程
- 809 篇 电子科学与技术（可...
- 613 篇 生物医学工程（可授...
- 477 篇 机械工程
- 474 篇 仪器科学与技术
- 453 篇 光学工程
- 302 篇 动力工程及工程热...
- 254 篇 生物工程
- 185 篇 交通运输工程
- 166 篇 航空宇航科学与技...
- 142 篇 网络空间安全
- 122 篇 材料科学与工程（可...
- 114 篇 石油与天然气工程
2,765 篇 理学
- 1,596 篇 数学
- 1,134 篇 物理学
- 406 篇 统计学（可授理学、...
- 294 篇 生物学
- 159 篇 系统科学
- 132 篇 化学
1,774 篇 医学
- 1,669 篇 临床医学
- 204 篇 基础医学(可授医学...
- 122 篇 药学(可授医学、理...
- 108 篇 特种医学
875 篇 管理学
- 608 篇 管理科学与工程(可...
- 321 篇 图书情报与档案管...
- 106 篇 工商管理
80 篇 法学
78 篇 农学
59 篇 经济学
38 篇 文学
28 篇 军事学
26 篇 教育学
19 篇 艺术学

主题

3,288 篇 data compression
1,744 篇 image coding
1,231 篇 image compressio...
1,137 篇 video compressio...
897 篇 point cloud comp...
646 篇 three-dimensiona...
571 篇 encoding
532 篇 compression
508 篇 transform coding
501 篇 decoding
450 篇 data mining
415 篇 compression algo...
369 篇 bandwidth
341 篇 video coding
268 篇 training
256 篇 streaming media
248 篇 computer science
243 篇 deep learning
238 篇 data models
232 篇 discrete cosine ...

机构

25 篇 tsinghua univers...
25 篇 univ chinese aca...
23 篇 ieee
23 篇 department of el...
22 篇 univ calif river...
22 篇 department of el...
22 篇 peng cheng lab p...
19 篇 peking univ peop...
19 篇 peking univ inst...
17 篇 university of ch...
17 篇 argonne natl lab...
16 篇 shanghai jiao to...
16 篇 univ sci & techn...
16 篇 signal and data ...
16 篇 hewlett packard ...
15 篇 harbin institute...
15 篇 department of au...
14 篇 penn state univ ...
14 篇 nanyang technolo...
14 篇 peng cheng labor...

作者

50 篇 w. kinsner
22 篇 liang xin
22 篇 tao dingwen
21 篇 xiong hongkai
18 篇 lukin vladimir
16 篇 klasky scott
16 篇 pratas diogo
15 篇 li li
15 篇 wang shiqi
12 篇 moffat a
12 篇 zhou wang
12 篇 ma zhan
11 篇 jin sian
11 篇 liu zhi
11 篇 wang wei
11 篇 shapira dana
10 篇 liu dong
10 篇 wang jing
10 篇 liu jinyang
10 篇 zorzi michele

语言

16,405 篇 英文
159 篇 其他
116 篇 中文
28 篇 土耳其文
3 篇 俄文
1 篇 西班牙文
1 篇 葡萄牙文

检索条件"任意字段=IEEE Data Compression Conference"

共 16713 条记录，以下是281-290 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

Unified data-Free compression: Pruning and Quantization without Fine-Tuning

Unified Data-Free Compression: Pruning and Quantization with...

引用

ieee/CVF International conference on Computer Vision (ICCV)

作者： Bai, Shipeng Chen, Jun Shen, Xintian Qian, Yixuan Liu, Yong Zhejiang Univ Coll Control Sci & Engn Hangzhou Peoples R China

ISBN: (纸本)9798350307184

Structured pruning and quantization are promising approaches for reducing the inference time and memory footprint of neural networks. However, most existing methods require the original training dataset to fine-tune the model. This not only brings heavy resource consumption but also is not possible for applications with sensitive or proprietary data due to privacy and security concerns. Therefore, a few data-free methods are proposed to address this problem, but they perform data-free pruning and quantization separately, which does not explore the complementarity of pruning and quantization. In this paper, we propose a novel framework named Unified data-Free compression(UDFC), which performs pruning and quantization simultaneously without any data and fine-tuning process. Specifically, UDFC starts with the assumption that the partial information of a damaged(e.g., pruned or quantized) channel can be preserved by a linear combination of other channels, and then derives the reconstruction form from the assumption to restore the information loss due to compression. Finally, we formulate the reconstruction error between the original network and its compressed network, and theoretically deduce the closed-form solution. We evaluate the UDFC on the large-scale image classification task and obtain significant improvements over various network architectures and compression methods. For example, we achieve a 20.54% accuracy improvement on ImageNet dataset compared to SOTA method with 30% pruning ratio and 6- bit quantization on ResNet-34. Code will be available at here.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A 1-TFLOPS/W, 28-nm Deep Neural Network Accelerator featuring Online compression and Decompression and BF16 Digital In-Memory-Computing Hardware 44

A 1-TFLOPS/W, 28-nm Deep Neural Network Accelerator featurin...

引用

Custom Integrated Circuits conference (CICC)

作者： Zhang, Bo Moon, Seunghyun Seok, Mingoo Columbia Univ New York NY 10027 USA

ISBN: (纸本)9798350394061

With the recent advances in deep neural networks (DNN), researchers have proposed various hardware accelerators. However, many neglected the energy consumption of off-chip memory access for weight and activation data, which can dominate the total energy consumption. To reduce off-chip data traffic, some works adopted aggressively quantized arithmetic, such as 1-4b fixed-point (FX1-FX4) or block-floating-point (BFP). However, they achieve only a limited computation precision, hurting DNN inference accuracy. © 2024 ieee.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Tensor compression and Reconstruction in Split DNN for Real-time Object Detection at the Edge

Tensor Compression and Reconstruction in Split DNN for Real-...

引用

ieee International Mediterranean conference on Communications and Networking (ieee MeditCom)

作者： Yu, Yenchia Evangelorato, Marco Chiasserini, Carla Fabiana Politecn Torino Turin Italy Univ Calif Irvine Irvine CA USA Chalmers Univ Technol Gothenburg Sweden

ISBN: (纸本)9798350309492;9798350309485

Computer vision applications for UAVs often rely on deep neural networks (DNNs) to increase prediction accuracy. As such DNNs crave for computational resources that can be hardly matched by those available at the UAVs, an emerging solution is to split the DNN into a low-complexity head model run at the UAV and a heavier tail run at the edge. This approach, however, comes at the cost of transmitting a large tensor data, hence of high bandwidth consumption, on the UAV-edge radio link. We tackle this problem by proposing the Compressed Tensor-based DNN split (CoTeD) framework, which executes tensor compression at the UAV and reconstruction at the edge, while conveniently trading off tensor compression with quality of the computer vision task output. When compared with the no-split case, CoTeD reduces the UAV computational burden by 50% w.r.t. performing inference at the UAV only, and the amount of transmitted data by over one order of magnitude w.r.t. running inference at the edge only. When compared to compressive sensing, JPEG-100, and the whole DNN run at the edge, CoTeD decreases the overall latency by (resp.) 95%, 75%, and 80%.

关键词： Edge computing UAVs Bandwidth utilization

来源：评论

学校读者我要写书评

暂无评论

DYNAMIC POINT CLOUD compression APPROACH USING HEXAHEDRON SEGMENTATION 30

DYNAMIC POINT CLOUD COMPRESSION APPROACH USING HEXAHEDRON SE...

引用

30th ieee International conference on Image Processing (ICIP)

作者： Tohidi, Faranak Paul, Manoranjan Charles Sturt Univ Sch Comp Math & Engn Bathurst NSW 2795 Australia

ISBN: (纸本)9781728198354

Video-based point cloud compression (V-PCC) is the state-of-the-art standard for compressing dynamic point clouds, a newly developed media format. However, real-time V-PCC applications are challenging because of excessive encoding complexity. This approach relies on converting 3D data into 2D frames using patch generation and refinement processes, which is one of the most computationally demanding components of V-PCC. This paper proposes a suitable-sized hexahedron and its projection plane by investigating different sizes of regular hexahedron partitioning and projection planes to convert 3D data to 2D without needing the complex patch generation and refinement processes while preserving more data and accuracy. The experimental results demonstrate that the chosen hexahedron can reduce the patch generation time of the V-PCC significantly. The proposed technique also minimises the point cloud losses, thus improving the image quality, and requires a smaller size of 2D maps than the V-PCC to further reduce the required bits for compression.

关键词： Dynamic point cloud compression 3D data V-PCC hexahedron partition

来源：评论

学校读者我要写书评

暂无评论

Attention-based Feature compression for CNN Inference Offloading in Edge Computing

Attention-based Feature Compression for CNN Inference Offloa...

引用

ieee International conference on Communications (ieee ICC)

作者： Li, Nan Iosifidis, Alexandros Zhang, Qi Aarhus Univ Dept Elect & Comp Engn DIGIT Aarhus Denmark

ISBN: (纸本)9781538674628

This paper studies the computational offloading of CNN inference in device-edge co-inference systems. Inspired by the emerging paradigm semantic communication, we propose a novel autoencoder-based CNN architecture (AECNN), for effective feature extraction at end-device. We design a feature compression module based on the channel attention method in CNN, to compress the intermediate data by selecting the most important features. To further reduce communication overhead, we can use entropy encoding to remove the statistical redundancy in the compressed data. At the receiver, we design a lightweight decoder to reconstruct the intermediate data through learning from the received compressed data to improve accuracy. To fasten the convergence, we use a step-by-step approach to train the neural networks obtained based on ResNet-50 architecture. Experimental results show that AECNN can compress the intermediate data by more than 256x with only about 4% accuracy loss, which outperforms the state-of-the-art work, BottleNet++. Compared to offloading inference task directly to edge server, AECNN can complete inference task earlier, in particular, under poor wireless channel condition, which highlights the effectiveness of AECNN in guaranteeing higher accuracy within time constraint.

关键词： CNN inference semantic communication feature compression Edge computing service reliability

来源：评论

学校读者我要写书评

暂无评论

ULTRA-LOW DELAY LOSSLESS compression OF HIGHER ORDER AMBISONICS 49

ULTRA-LOW DELAY LOSSLESS COMPRESSION OF HIGHER ORDER AMBISON...

引用

49th ieee International conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Namazi, Mahmoud Rose, Kenneth Univ Calif Santa Barbara Santa Barbara CA 93106 USA

ISBN: (纸本)9798350344868;9798350344851

Coding delay has emerged as a significant obstacle to wider adoption of Higher Order Ambisonics (HOA), a promising flexible format employed in representation and coding of spatial audio. Low delay is critical in many applications, such as virtual reality, gaming, and online music performance, a good subset of which further require lossless compression. However, major advances in spatial audio codec design often sacrifice low delay in order to maximize compression gains. We present a new lossless, ultra-low delay HOA codec, optimized to take advantage of both the spatial and temporal redundancies in HOA data, while maintaining an extremely low delay. The presented codec has an algorithmic delay of one single sample, at the cost of increased computational complexity. This codec outperforms the state-of-the-art for lossless, low-delay HOA coding, both in terms of delay and bitrate.

关键词： audio coding ambisonics spatial audio lossless compression low delay coding

来源：评论

学校读者我要写书评

暂无评论

Revisiting data Augmentation in Model compression: An Empirical and Comprehensive Study

Revisiting Data Augmentation in Model Compression: An Empiri...

引用

International Joint conference on Neural Networks (IJCNN)

作者： Yu, Muzhou Zhang, Linfeng Ma, Kaisheng Xi An Jiao Tong Univ Xian Peoples R China Tsinghua Univ Beijing Peoples R China

ISBN: (纸本)9781665488679

The excellent performance of deep neural networks is usually accompanied by a large number of parameters and computations, which have limited their usage on the resource-limited edge devices. To address this issue, abundant methods such as pruning, quantization and knowledge distillation have been proposed to compress neural networks and achieved significant breakthroughs. However, most of these compression methods focus on the architecture or the training method of neural networks but ignore the influence from data augmentation. In this paper, we revisit the usage of data augmentation in model compression and give a comprehensive study on the relation between model sizes and their optimal data augmentation policy. To sum up, we mainly have the following three observations: (A) Models in different sizes prefer data augmentation with different magnitudes. Hence, in iterative pruning, data augmentation with varying magnitudes leads to better performance than data augmentation with a consistent magnitude. (B) data augmentation with a high magnitude may significantly improve the performance of large models but harm the performance of small models. Fortunately, small models can still benefit from strong data augmentations by firstly learning them with "additional parameters" and then discard these "additional parameters" during inference. (C) The prediction of a pre-trained large model can be utilized to measure the difficulty of data augmentation. Thus it can be utilized as a criterion to design better data augmentation policies. We hope this paper may promote more research on the usage of data augmentation in model compression.

关键词： data augmentation model compression

来源：评论

学校读者我要写书评

暂无评论

Complexity Reduction of Neural Networks for Nonlinear Fourier Transform in Optical Transmission Systems 1

Complexity Reduction of Neural Networks for Nonlinear Fourie...

引用

1st ieee International conference on Machine Learning for Communication and Networking (ieee ICMLCN)

作者： Bogdanov, Stepan Freire, Pedro J. Prilepsky, Jaroslaw E. Aston Univ AIPT Birmingham England

ISBN: (纸本)9798350343205;9798350343199

The nonlinear Fourier transform (NFT) method was proposed a decade ago as a tool for the mitigation of nonlinear impairments in fiber-optic communication systems. Within the NFT-based transmission methods, we use the parameters of the so-called nonlinear modes of non-decaying signals as data carriers. Naturally, to make the NFT transmission more practical, we need an efficient low-complexity method for performing the NFT processing of our signals at the transmitter and receiver. Utilizing neural network models has recently become a popular direction in signal processing, including studies related to the application of neural networks for NFT computations. Complexity reduction (model compression) is an important step for the practical implementation of neural networks in signal-processing applications. In this work, we develop a weight clustering compression technique applied to the convolutional neural network (CNN) performing a direct NFT operation. This CNN is used to retrieve the phases of non-decaying nonlinear solutions (nonlinear modes) to the nonlinear Shrodinger equation. We found that satisfactory model performance is retained even when the number of clusters falls to four. When utilized at the receiver of a communication system, our compressed model demonstrates the data restoration quality below the 7% forward error correction (FEC) threshold with the number of clusters starting from six where compression of the model reaches 99%.

关键词： convolutional neural network model compression nonlinear Fourier transform weight clustering

来源：评论

学校读者我要写书评

暂无评论

Bitmap Dynamic Header Optimization for Overhead Reduction in IP Services for Narrowband Channels 21

Bitmap Dynamic Header Optimization for Overhead Reduction in...

引用

21st ieee International conference on Mobile Ad-Hoc and Smart Systems (MASS)

作者： Dhaka, Manoj Kumar Peddoju, Sateesh K. Sharma, S. C. DRDO Def Elect Applicat Lab Dehra Dun Uttarakhand India Indian Inst Technol Roorkee Uttar Pradesh India

ISBN: (纸本)9798350363999;9798350364002

When carrying IP traffic over low-bandwidth narrowband links, the IP header consumes a significant amount of bandwidth. In many deployments, tactical networks possess limited link capacity. Substantial reduction in the transmission overhead is possible by suppressing the repetition of static information, and by synchronizing the sender and receiver for reproducing the information. Instead of sending all the fields of the header with every packet, information of some selected fields can be transferred based on the consideration of how these fields vary over the period of the data transfer. The approach can result in a considerable reduction in header overhead thereby improving the bandwidth utilization. The paper introduces the concept of dynamism in the compression process by utilizing the bit-map pattern. To mitigate the loss of primary information, an acknowledgement-based handshake mechanism has been introduced which is supplemented by synchronization maintenance. The results evident the improved bandwidth utilization, less overhead, and header compression efficiency of 50% to 90% which is revitalizing for narrowband channels and low bandwidth MANET environment.

关键词： bandwidth bitmap header compression IP services narrowband links

来源：评论

学校读者我要写书评

暂无评论

Condar: Context-Aware Distributed Dynamic Object Detection on Radar data

Condar: Context-Aware Distributed Dynamic Object Detection o...

引用

2024 Military Communications conference

作者： Harshbarger, Ian Zejda, Colin Levorato, Marco Univ Calif Irvine Irvine CA 92697 USA

ISBN: (纸本)9798350374247;9798350374230

Deep neural networks are an established class of algorithms widely used for real-time data analysis in computer vision-based mobile applications. However, on one hand the constrained resources - computing power, energy reservoir and memory - available to mobile devices clash with the complexity of high-performance neural models. On the other hand, offloading the execution of the computing task to a wirelessly connected edge/cloud server requires the transfer of information rich signals over volatile and capacity constrained channels. To address this issue, the Split Computing (SC) paradigm has been proposed. In SC frameworks, a baseline DNN model is used to create two models, namely head and tail models, that are executed at a mobile device and edge/cloud server, respectively. In this work, a SC paradigm - Condar - is presented that introduces several core design and conceptual innovations: (i) we target object detection on radar data rather than 2D RGB images;(ii) we design encoders/decoder structures that are specialized to the specific operating environment;(iii) we develop a multi-branched model that can dynamically select the best performing encoder while executing the main model. Our design is based on the CenterNet model with a split ResNet50 backbone trained on the RADIATE dataset's radar images of resolution 1152x1152. Our results show that Condar has performance analogous to that of a Generalized CenterNet Model with a difference of 4.41%, 1.72%, and 2.18% on mAP@50, mAP@75, and mAP@COCO respectively, while simultaneously performing compression on the features of the data to 15.5KB. This datasize is smaller than lowest possible compression data size of JPEG compression at 21KB, which results in a considerable performance degradation.

关键词： Dynamic Deep Neural Networks Edge Computing Split Computing Radar Object Detection

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共500页 << < 25 26 27 28 29 30 31 32 33 34 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

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：