内容简介：同时提供每月大咖直播分享、真实项目需求对接、干货资讯汇总，行业技术交流*延伸阅读
加入极市 专业CV交流群，与 6000+来自腾讯，华为，百度，北大，清华，中科院 等名企名校视觉开发者互动交流！更有机会与 李开复老师 等大牛群内互动！
同时提供每月大咖直播分享、真实项目需求对接、干货资讯汇总，行业技术交流 。 点击文末“ 阅读原文 ”立刻申请入群~
作者 | Mars_WH
来源 |
blog.csdn.net/hw5226349/article/details/84888416
原文 | http://bbs.cvmart.net/topics/352
深度学习（Deep Learning）因其计算复杂度或参数冗余，在一些场景和设备上限制了相应的模型部署，需要借助模型压缩、优化加速、异构计算等方法突破瓶颈。其中模型压缩算法能够有效降低参数冗余，从而减少存储占用、通信带宽和计算复杂度，有助于深度学习的应用部署，本文汇总了近几年模型压缩方面的相关研究paper，欢迎收藏阅读~
结构structure
Searching for MobileNetV3
arxiv:https://arxiv.org/abs/1905.02244v1
中文解读： 重磅！MobileNetV3 来了！
[BMVC2018] IGCV3: Interleaved Low-Rank Group Convolutions for Efficient Deep Neural Networks
https://arxiv.org/abs/1806.00178
github: https://github.com/homles11/IGCV3
[CVPR2018] IGCV2: Interleaved Structured Sparse Convolutional Neural Networks
arxiv: https://arxiv.org/abs/1804.06202
[CVPR2018] MobileNetV2: Inverted Residuals and Linear Bottlenecks
arxiv: https://arxiv.org/abs/1801.04381
github: https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet
[ECCV2018] ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design
arxiv: https://arxiv.org/abs/1807.11164
量化quantization
intro:二值网络
https://arxiv.org/abs/1602.02830github: https://github.com/MatthieuCourbariaux/BinaryNet
https://github.com/itayhubara/BinaryNet
intro:二值网络
http://www.idi.ntnu.no/~yamanu/2017-fpga-finn-preprint.pdfgithub: https://github.com/Xilinx/FINN
intro:低bit位
https://arxiv.org/abs/1606.06160github: https://github.com/tensorpack/tensorpack/tree/master/examples/DoReFa-Net
intro:darknet团队出品
https://arxiv.org/abs/1603.05279github: https://github.com/allenai/XNOR-Net
[CVPR2016] Ternary Weight Networks
arxiv: https://arxiv.org/abs/1605.04711
github: https://github.com/fengfu-chris/caffe-twns
Google出品
https://arxiv.org/abs/1712.05877github: https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/quantize
intro:QNNs
https://arxiv.org/abs/1609.07061github: https://github.com/peisuke/qnn
Two-Step Quantization for Low-bit Neural Networks
paper: http://openaccess.thecvf.com/content_cvpr_2018/papers/Wang_Two-Step_Quantization_for_CVPR_2018_paper.pdf
剪枝pruning
通道裁剪channel pruning
[NIPS2018] Discrimination-aware Channel Pruning for Deep Neural Networks
arxiv: https://arxiv.org/abs/1810.11809
github: https://github.com/Tencent/PocketFlow支持DisChnPrunedLearner
intro:Lasso回归
https://arxiv.org/abs/1707.06168github: https://github.com/yihui-he/channel-pruning
intro:自动学习优化
https://arxiv.org/abs/1802.03494https://www.jiqizhixin.com/articles/AutoML-for-Model-Compression-and-Acceleration-on-Mobile-Devices论文翻译
github: https://github.com/Tencent/PocketFlow
[ICCV2017] Learning Efficient Convolutional Networks through Network Slimming
intro:Zhuang Liu
https://arxiv.org/abs/1708.06519github: https://github.com/Eric-mingjie/network-slimming
https://github.com/foolwood/pytorch-slimming
[ICLR2018] Rethinking the Smaller-Norm-Less-Informative Assumption in Channel Pruning of Convolution Layers
arxiv: https://arxiv.org/abs/1802.00124
github:[PyTorch] https://github.com/jack-willturner/batchnorm-pruning
[TensorFlow] https://github.com/bobye/batchnorm_prune
[CVPR2017] NISP: Pruning Networks using Neuron Importance Score Propagation
arxiv: https://arxiv.org/abs/1711.05908
[ICCV2017] ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression
web: http://lamda.nju.edu.cn/luojh/project/ThiNet_ICCV17/ThiNet_ICCV17_CN.html
github: https://github.com/Roll920/ThiNet
https://github.com/Roll920/ThiNet_Code
稀疏sparsity
intro: Uber
https://arxiv.org/abs/1801.02108github: https://github.com/uber/sbnet
intro:稀疏
https://arxiv.org/abs/1710.01878github: https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/model_pruning
intro:Facebook
https://arxiv.org/abs/1706.01307github: https://github.com/facebookresearch/SparseConvNet
融合fusion
蒸馏distillation
intro:Hinton出品
https://arxiv.org/abs/1503.02531github: https://github.com/peterliht/knowledge-distillation-pytorch
综合comprehensive
intro:开创先河
https://arxiv.org/abs/1510.00149github: https://github.com/songhan
Model Distillation with Knowledge Transfer from Face Classification to Alignment and Verificationintro:实验比较多，适合工程化
arxiv:
https://arxiv.org/abs/1709.02929
持续更新中，欢迎关注
*延伸阅读
-
Facebook发布PyTorch 1.1，开源AI模型优化简化工具BoTorch & Ax
-
不用重新训练，直接将现有模型转换为 MobileNet
-
TVM: Deep Learning模型的优化编译器(强烈推荐, 附踩坑记录)
-
论文综述：当前深度神经网络模型压缩和加速方法速览
点击左下角 “ 阅读原文 ”， 即可申请加入极市 目标跟踪、目标检测、工业检测、人脸方向、视觉竞赛等技术交流群， 更有每月大咖直播分享、真实项目需求对接、干货资讯汇总，行业技术交流， 一起来让思想之光照的更远吧~
觉得有用麻烦给个在看啦~
以上就是本文的全部内容，希望本文的内容对大家的学习或者工作能带来一定的帮助，也希望大家多多支持 码农网
猜你喜欢:
本站部分资源来源于网络，本站转载出于传递更多信息之目的，版权归原作者或者来源机构所有，如转载稿涉及版权问题，请联系我们。
The Haskell School of Music
Paul Hudak、Donya Quick / Cambridge University Press / 2018-10-4 / GBP 42.99
This book teaches functional programming through creative applications in music and sound synthesis. Readers will learn the Haskell programming language and explore numerous ways to create music and d......一起来看看 《The Haskell School of Music》 这本书的介绍吧!
