Compression of convolutional neural networks: A short survey

机译：卷积神经网络的压缩：简短调查

获取原文

获取原文并翻译 | 示例

页面导航

摘要
著录项
相似文献
相关主题

摘要

Nowadays, convolutional neural networks (CNN) are considered as the state-of-the-art algorithms for various tasks, especially for image classification and recognition. Because of that, more and more attention is aimed towards implementation of CNNs on embedded systems. Main obstacles for implementing CNNs on embedded systems are their large model size and large number of operations needed for inference. In order to surpass these obstacles, algorithms for CNN compression tend to lower model size and number of operations needed for inference. In this paper we review the state-of-the-art in CNN compression. To this end we divided all approaches for CNN compression into three groups: precision reduction, network pruning and design of compact network architectures. After presenting the main approaches in each group we conclude that the future CNN compression algorithms should be co-designed with hardware which will process deep learning algorithms.

机译：如今，卷积神经网络（CNN）被认为是用于各种任务（尤其是图像分类和识别）的最新算法。因此，越来越多的注意力旨在在嵌入式系统上实现CNN。在嵌入式系统上实现CNN的主要障碍是其大型模型和推理所需的大量操作。为了克服这些障碍，用于CNN压缩的算法倾向于降低模型大小和推理所需的操作数量。在本文中，我们回顾了CNN压缩的最新技术。为此，我们将CNN压缩的所有方法分为三类：精度降低，网络修剪和紧凑网络体系结构设计。在介绍了每个小组的主要方法之后，我们得出结论，未来的CNN压缩算法应与将处理深度学习算法的硬件共同设计。

著录项

来源
《2018 17th International Symposium INFOTEH-JAHORINA》|2018年|1-6|共6页
会议地点 East Sarajevo(BA)
作者
Ratko Pilipović; Patricio Bulić; Vladimir Risojević;
展开▼
作者单位

Faculty of Computer and Information Science, University in Ljubljana, Ljubljana, Republic Slovenia;

Faculty of Computer and Information Science, University in Ljubljana, Ljubljana, Republic Slovenia;

Faculty of Electrical Engineering, University in Banja Luka, Banja Luka, Bosnia and Herzegovina;

展开▼
会议组织
原文格式 PDF
正文语种 eng
中图分类
关键词
convolution; embedded systems; feedforward neural nets;

机译：卷积;嵌入式系统;前馈神经网络;;
入库时间 2022-08-26 14:05:04

相似文献

外文文献
中文文献
专利

1. Groundwater level forecasting with artificial neural networks: a comparison of long short-term memory (LSTM), convolutional neural networks (CNNs), and non-linear autoregressive networks with exogenous input (NARX) [J] . Wunsch Andreas, Liesch Tanja, Broda Stefan Hydrology and Earth System Sciences Discussions . 2021,第3期

机译：与人工神经网络的地下水位预测：具有外源输入（NARX）的长短期记忆（LSTM），卷积神经网络（CNNS）和非线性自回归网络的比较
2. Groundwater level forecasting with artificial neural networks: a comparison of long short-term memory (LSTM), convolutional neural networks (CNNs), and non-linear autoregressive networks with exogenous input (NARX) [J] . Wunsch Andreas, Liesch Tanja, Broda Stefan Hydrology and Earth System Sciences . 2021,第3期

机译：与人工神经网络的地下水位预测：具有外源输入（NARX）的长短期记忆（LSTM），卷积神经网络（CNNS）和非线性自回归网络的比较
3. Cascade convolutional neural network-long short-term memory recurrent neural networks for automatic tonal and nontonal preclassification-based Indian language identification [J] . China Bhanja Chuya, Laskar Mohammad A., Laskar Rabul H. Expert Systems . 2020,第5期

机译：级联卷积神经网络长短期内存经常性神经网络，用于自动色调和非统计学预分配的印度语言识别
4. Compression of convolutional neural networks: A short survey [C] . Ratko Pilipovi?, Patricio Buli?, Vladimir Risojevi? International Symposium INFOTEH-JAHORINA . 2018

机译：压缩卷积神经网络：短暂的调查
5. Framework for Acceleration and Compression of Convolutional Neural Networks [D] . Misko, Joshua 2018

机译：卷积神经网络加速和压缩框架
6. Recognition of Abnormal Chest Compression Depth Using One-Dimensional Convolutional Neural Networks [O] . Liang Zhao, Yu Bao, Yu Zhang, 2021

机译：用一维卷积神经网络识别异常胸部压缩深度
7. How convolutional neural networks see the world --- A survey of convolutional neural network visualization methods [O] . Zhuwei Qin, Fuxun Yu, Chenchen Liu, 2018

机译：卷积神经网络如何看世界---卷积神经网络可视化方法的调查

Compression of convolutional neural networks: A short survey

摘要

著录项

相似文献

相关主题

期刊订阅