SlideShare una empresa de Scribd logo
1 de 17
Descargar para leer sin conexión
CNN Structure: from LeNet to ShuffleNet
Dalin Zhang
School of CSE, UNSW
17/Jul/2017
Preliminary
Local receptive fields
Shared weights and biases
Multiple feature maps
Subsampling maps
LeNet: Hello World!
Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner,
Gradient-based learning applied to document recognition, Proc. IEEE 86(11):
2278–2324, 1998.
C(5x5)-P(2x2) pair repeat
Average pooling
Sigmoid or tanh activation function
ILSVRC
ImageNet Large Scale Visual Recognition Challenge (ILSVRC)
More than 1.2 Million Images 1000 classes
Impressive new CNN structures from ILSVRC
www.image-
net.org/challenges/LSVRC/
AlexNet: ILSVRC 2012 winner
C(11x11)P-C(5x5)P-C(3x3)-C(3x3)-C(3x3)P
Max pooling
Relu activation function
8 layers
A. Krizhevsky, I. Sutskever, and G. Hinton,
ImageNet Classification with Deep Convolutional Neural Networks, NIPS
2012
VGGNet: ILSVRC 2014 2nd
All convolutional layer kernels are of size 3x3
MaxPooling of size 2x2 is done after 2 or 3 layers of convolutions
Pooling stride is 2
Stacking building blocks of the same shape
K. Simonyan and A. Zisserman,
Very Deep Convolutional Networks for Large-Scale Image Recognition, ICLR
2015
GoogleNet: ILSVRC 2014 Winner
Let the network choose the kernel size itself
Pointwise convolution (1x1 convolution) reduce parameters
22 layers
C. Szegedy et al.,
Going deeper with convolutions, CVPR 2015
ResNet: ILSVRC 2015 Winner
Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun,
Deep Residual Learning for Image Recognition, CVPR 2016 (Best Paper)
ResNet: 152
layers
ResNet: ILSVRC 2015 Winner
Introduce skip connections
Pointwise convolution reduce and restore feature maps
152 layers, top-5 error rate 3.57% vs. 5.1% of human expert
Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun,
Deep Residual Learning for Image Recognition, CVPR 2016 (Best Paper)
Directly performing 3x3 convolutions:
Parameters: 256x256x3x3 ~ 600K
Residual module structure:
Parameters:
64x256x1x1 ~ 16K
64x64x3x3 ~ 36K
256x64x1x1 ~ 16K
Total ~70K
ResNet: ILSVRC 2015 Winner
Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun,
Deep Residual Learning for Image Recognition, CVPR 2016 (Best Paper)
Problem:
with the network depth increasing, accuracy gets saturated (which might be
unsurprising) and then degrades rapidly.
Deeper network is not easy to optimize.
Cause:
In some cases some neuron can “die”(output zero) in the training and
become ineffective/useless. This can cause information loss, sometimes
very important information.
Solution:
Skip connections carry important information in the previous layer to the
next layers.
Xception: Depthwise Separable Convolutions
François Chollet
Xception: Deep Learning with Depthwise Separable Convolution (2017 Apr)
Important Hypothesis:
The mapping of cross-channels correlations and spatial correlations in the
feature maps of convolutional neural networks can be entirely decoupled.
output
input
Xception: Depthwise Separable Convolutions
François Chollet
Xception: Deep Learning with Depthwise Separable Convolution (2017 Apr)
ResNeXt: Group Convolutions ILSVRC 2016 2nd
Saining Xie, Ross Girshick, Piotr Dollar, Zhuowen Tu, Kaiming He
Aggregated Residual Transformations for Deep Neural Networks (2017 Apr)
Introduce group convolution to the ResNet unit, thus introduce
a new dimension “cardinality” (the number of groups) to
ResNet.
ResNeXt: Group Convolutions ILSVRC 2016 2nd
More clear case
Group convolution reduce the complexity compared to the similar ResNet
structure. Gain better performance at the same complexity
Saining Xie, Ross Girshick, Piotr Dollar, Zhuowen Tu, Kaiming He
Aggregated Residual Transformations for Deep Neural Networks (2017 Apr)
ShuffleNeXt: pointwise group conv+channel shuffle
Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin and Jian Sun
ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
Channel shuffle: help information flow across feature maps
(B, g x n, H, W) – reshape(B, g, n, H, W) – transpose(B, n, g, H, W) –
reshape(B, g, n, H, W)
ShuffleNeXt: pointwise group conv+channel shuffle
Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin and Jian Sun
ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
Pointwise group convolution:
Reduce complexity allowing more feature maps, especially important to small
networks
Summary
 Stack simple structures
 Skip connection
 Pointwise convolution
 Depthwise convolution
 Group convolution
 Channel shuffle

Más contenido relacionado

La actualidad más candente

Conditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN DecodersConditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN Decoderssuga93
 
Convolutional neural networks
Convolutional neural networksConvolutional neural networks
Convolutional neural networksSlobodan Blazeski
 
[DL輪読会]Learning Visible Connectivity Dynamics for Cloth Smoothing (CoRL2021)
[DL輪読会]Learning Visible Connectivity Dynamics for Cloth Smoothing (CoRL2021)[DL輪読会]Learning Visible Connectivity Dynamics for Cloth Smoothing (CoRL2021)
[DL輪読会]Learning Visible Connectivity Dynamics for Cloth Smoothing (CoRL2021)Deep Learning JP
 
Visualizaing and understanding convolutional networks
Visualizaing and understanding convolutional networksVisualizaing and understanding convolutional networks
Visualizaing and understanding convolutional networksSungminYou
 
Convolutional Neural Networks - Xavier Giro - UPC TelecomBCN Barcelona 2020
Convolutional Neural Networks - Xavier Giro - UPC TelecomBCN Barcelona 2020Convolutional Neural Networks - Xavier Giro - UPC TelecomBCN Barcelona 2020
Convolutional Neural Networks - Xavier Giro - UPC TelecomBCN Barcelona 2020Universitat Politècnica de Catalunya
 
Unsupervised Learning (D2L6 2017 UPC Deep Learning for Computer Vision)
Unsupervised Learning (D2L6 2017 UPC Deep Learning for Computer Vision)Unsupervised Learning (D2L6 2017 UPC Deep Learning for Computer Vision)
Unsupervised Learning (D2L6 2017 UPC Deep Learning for Computer Vision)Universitat Politècnica de Catalunya
 
Problems with CNNs and Introduction to capsule neural networks
Problems with CNNs and Introduction to capsule neural networksProblems with CNNs and Introduction to capsule neural networks
Problems with CNNs and Introduction to capsule neural networksVipul Vaibhaw
 
Convolutional Neural Network (CNN)
Convolutional Neural Network (CNN)Convolutional Neural Network (CNN)
Convolutional Neural Network (CNN)Muhammad Haroon
 
Transformer 動向調査 in 画像認識
Transformer 動向調査 in 画像認識Transformer 動向調査 in 画像認識
Transformer 動向調査 in 画像認識Kazuki Maeno
 
Visualizing and Understanding Convolutional Networks
Visualizing and Understanding Convolutional NetworksVisualizing and Understanding Convolutional Networks
Visualizing and Understanding Convolutional NetworksWilly Marroquin (WillyDevNET)
 
[PR12] PR-063: Peephole predicting network performance before training
[PR12] PR-063: Peephole predicting network performance before training[PR12] PR-063: Peephole predicting network performance before training
[PR12] PR-063: Peephole predicting network performance before trainingTaegyun Jeon
 
Deep convnets for global recognition (Master in Computer Vision Barcelona 2016)
Deep convnets for global recognition (Master in Computer Vision Barcelona 2016)Deep convnets for global recognition (Master in Computer Vision Barcelona 2016)
Deep convnets for global recognition (Master in Computer Vision Barcelona 2016)Universitat Politècnica de Catalunya
 
capsule network
capsule networkcapsule network
capsule network민기 정
 
Deep Learning for Computer Vision: Transfer Learning and Domain Adaptation (U...
Deep Learning for Computer Vision: Transfer Learning and Domain Adaptation (U...Deep Learning for Computer Vision: Transfer Learning and Domain Adaptation (U...
Deep Learning for Computer Vision: Transfer Learning and Domain Adaptation (U...Universitat Politècnica de Catalunya
 
Visualization of Deep Learning Models (D1L6 2017 UPC Deep Learning for Comput...
Visualization of Deep Learning Models (D1L6 2017 UPC Deep Learning for Comput...Visualization of Deep Learning Models (D1L6 2017 UPC Deep Learning for Comput...
Visualization of Deep Learning Models (D1L6 2017 UPC Deep Learning for Comput...Universitat Politècnica de Catalunya
 
Efficient Neural Network Architecture for Image Classfication
Efficient Neural Network Architecture for Image ClassficationEfficient Neural Network Architecture for Image Classfication
Efficient Neural Network Architecture for Image ClassficationYogendra Tamang
 

La actualidad más candente (20)

Deep 3D Visual Analysis - Javier Ruiz-Hidalgo - UPC Barcelona 2017
Deep 3D Visual Analysis - Javier Ruiz-Hidalgo - UPC Barcelona 2017Deep 3D Visual Analysis - Javier Ruiz-Hidalgo - UPC Barcelona 2017
Deep 3D Visual Analysis - Javier Ruiz-Hidalgo - UPC Barcelona 2017
 
Conditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN DecodersConditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN Decoders
 
Convolutional neural networks
Convolutional neural networksConvolutional neural networks
Convolutional neural networks
 
Deep Visual Saliency - Kevin McGuinness - UPC Barcelona 2017
Deep Visual Saliency - Kevin McGuinness - UPC Barcelona 2017Deep Visual Saliency - Kevin McGuinness - UPC Barcelona 2017
Deep Visual Saliency - Kevin McGuinness - UPC Barcelona 2017
 
Mnist report
Mnist reportMnist report
Mnist report
 
[DL輪読会]Learning Visible Connectivity Dynamics for Cloth Smoothing (CoRL2021)
[DL輪読会]Learning Visible Connectivity Dynamics for Cloth Smoothing (CoRL2021)[DL輪読会]Learning Visible Connectivity Dynamics for Cloth Smoothing (CoRL2021)
[DL輪読会]Learning Visible Connectivity Dynamics for Cloth Smoothing (CoRL2021)
 
Visualizaing and understanding convolutional networks
Visualizaing and understanding convolutional networksVisualizaing and understanding convolutional networks
Visualizaing and understanding convolutional networks
 
Convolutional Neural Networks - Xavier Giro - UPC TelecomBCN Barcelona 2020
Convolutional Neural Networks - Xavier Giro - UPC TelecomBCN Barcelona 2020Convolutional Neural Networks - Xavier Giro - UPC TelecomBCN Barcelona 2020
Convolutional Neural Networks - Xavier Giro - UPC TelecomBCN Barcelona 2020
 
Intepretability / Explainable AI for Deep Neural Networks
Intepretability / Explainable AI for Deep Neural NetworksIntepretability / Explainable AI for Deep Neural Networks
Intepretability / Explainable AI for Deep Neural Networks
 
Unsupervised Learning (D2L6 2017 UPC Deep Learning for Computer Vision)
Unsupervised Learning (D2L6 2017 UPC Deep Learning for Computer Vision)Unsupervised Learning (D2L6 2017 UPC Deep Learning for Computer Vision)
Unsupervised Learning (D2L6 2017 UPC Deep Learning for Computer Vision)
 
Problems with CNNs and Introduction to capsule neural networks
Problems with CNNs and Introduction to capsule neural networksProblems with CNNs and Introduction to capsule neural networks
Problems with CNNs and Introduction to capsule neural networks
 
Convolutional Neural Network (CNN)
Convolutional Neural Network (CNN)Convolutional Neural Network (CNN)
Convolutional Neural Network (CNN)
 
Transformer 動向調査 in 画像認識
Transformer 動向調査 in 画像認識Transformer 動向調査 in 画像認識
Transformer 動向調査 in 画像認識
 
Visualizing and Understanding Convolutional Networks
Visualizing and Understanding Convolutional NetworksVisualizing and Understanding Convolutional Networks
Visualizing and Understanding Convolutional Networks
 
[PR12] PR-063: Peephole predicting network performance before training
[PR12] PR-063: Peephole predicting network performance before training[PR12] PR-063: Peephole predicting network performance before training
[PR12] PR-063: Peephole predicting network performance before training
 
Deep convnets for global recognition (Master in Computer Vision Barcelona 2016)
Deep convnets for global recognition (Master in Computer Vision Barcelona 2016)Deep convnets for global recognition (Master in Computer Vision Barcelona 2016)
Deep convnets for global recognition (Master in Computer Vision Barcelona 2016)
 
capsule network
capsule networkcapsule network
capsule network
 
Deep Learning for Computer Vision: Transfer Learning and Domain Adaptation (U...
Deep Learning for Computer Vision: Transfer Learning and Domain Adaptation (U...Deep Learning for Computer Vision: Transfer Learning and Domain Adaptation (U...
Deep Learning for Computer Vision: Transfer Learning and Domain Adaptation (U...
 
Visualization of Deep Learning Models (D1L6 2017 UPC Deep Learning for Comput...
Visualization of Deep Learning Models (D1L6 2017 UPC Deep Learning for Comput...Visualization of Deep Learning Models (D1L6 2017 UPC Deep Learning for Comput...
Visualization of Deep Learning Models (D1L6 2017 UPC Deep Learning for Comput...
 
Efficient Neural Network Architecture for Image Classfication
Efficient Neural Network Architecture for Image ClassficationEfficient Neural Network Architecture for Image Classfication
Efficient Neural Network Architecture for Image Classfication
 

Similar a CNN Structure: From LeNet to ShuffleNet

UNetEliyaLaialy (2).pptx
UNetEliyaLaialy (2).pptxUNetEliyaLaialy (2).pptx
UNetEliyaLaialy (2).pptxNoorUlHaq47
 
#4 Convolutional Neural Networks for Natural Language Processing
#4 Convolutional Neural Networks for Natural Language Processing#4 Convolutional Neural Networks for Natural Language Processing
#4 Convolutional Neural Networks for Natural Language ProcessingBerlin Language Technology
 
Convolutional neural network from VGG to DenseNet
Convolutional neural network from VGG to DenseNetConvolutional neural network from VGG to DenseNet
Convolutional neural network from VGG to DenseNetSungminYou
 
Deep learning lecture - part 1 (basics, CNN)
Deep learning lecture - part 1 (basics, CNN)Deep learning lecture - part 1 (basics, CNN)
Deep learning lecture - part 1 (basics, CNN)SungminYou
 
Recent advances of AI for medical imaging : Engineering perspectives
Recent advances of AI for medical imaging : Engineering perspectivesRecent advances of AI for medical imaging : Engineering perspectives
Recent advances of AI for medical imaging : Engineering perspectivesNamkug Kim
 
Details of Lazy Deep Learning for Images Recognition in ZZ Photo app
Details of Lazy Deep Learning for Images Recognition in ZZ Photo appDetails of Lazy Deep Learning for Images Recognition in ZZ Photo app
Details of Lazy Deep Learning for Images Recognition in ZZ Photo appPAY2 YOU
 
20141003.journal club
20141003.journal club20141003.journal club
20141003.journal clubHayaru SHOUNO
 
[PR12] Inception and Xception - Jaejun Yoo
[PR12] Inception and Xception - Jaejun Yoo[PR12] Inception and Xception - Jaejun Yoo
[PR12] Inception and Xception - Jaejun YooJaeJun Yoo
 
(Research Note) Delving deeper into convolutional neural networks for camera ...
(Research Note) Delving deeper into convolutional neural networks for camera ...(Research Note) Delving deeper into convolutional neural networks for camera ...
(Research Note) Delving deeper into convolutional neural networks for camera ...Jacky Liu
 
Learning Sparse Neural Networksvia Sensitivity-Driven Regularization
Learning Sparse Neural Networksvia Sensitivity-Driven RegularizationLearning Sparse Neural Networksvia Sensitivity-Driven Regularization
Learning Sparse Neural Networksvia Sensitivity-Driven RegularizationEnzo Tartaglione
 
Modeling perceptual similarity and shift invariance in deep networks
Modeling perceptual similarity and shift invariance in deep networksModeling perceptual similarity and shift invariance in deep networks
Modeling perceptual similarity and shift invariance in deep networksNAVER Engineering
 
AI&BigData Lab. Артем Чернодуб "Распознавание изображений методом Lazy Deep ...
AI&BigData Lab. Артем Чернодуб  "Распознавание изображений методом Lazy Deep ...AI&BigData Lab. Артем Чернодуб  "Распознавание изображений методом Lazy Deep ...
AI&BigData Lab. Артем Чернодуб "Распознавание изображений методом Lazy Deep ...GeeksLab Odessa
 
ct_meeting_final_jcy (1).pdf
ct_meeting_final_jcy (1).pdfct_meeting_final_jcy (1).pdf
ct_meeting_final_jcy (1).pdfssuser2c7393
 

Similar a CNN Structure: From LeNet to ShuffleNet (20)

UNetEliyaLaialy (2).pptx
UNetEliyaLaialy (2).pptxUNetEliyaLaialy (2).pptx
UNetEliyaLaialy (2).pptx
 
#4 Convolutional Neural Networks for Natural Language Processing
#4 Convolutional Neural Networks for Natural Language Processing#4 Convolutional Neural Networks for Natural Language Processing
#4 Convolutional Neural Networks for Natural Language Processing
 
Convolutional neural network from VGG to DenseNet
Convolutional neural network from VGG to DenseNetConvolutional neural network from VGG to DenseNet
Convolutional neural network from VGG to DenseNet
 
Deep learning lecture - part 1 (basics, CNN)
Deep learning lecture - part 1 (basics, CNN)Deep learning lecture - part 1 (basics, CNN)
Deep learning lecture - part 1 (basics, CNN)
 
CNN
CNNCNN
CNN
 
Recent advances of AI for medical imaging : Engineering perspectives
Recent advances of AI for medical imaging : Engineering perspectivesRecent advances of AI for medical imaging : Engineering perspectives
Recent advances of AI for medical imaging : Engineering perspectives
 
Spectral convnets
Spectral convnetsSpectral convnets
Spectral convnets
 
Deep Learning
Deep LearningDeep Learning
Deep Learning
 
Deep learning and computer vision
Deep learning and computer visionDeep learning and computer vision
Deep learning and computer vision
 
Details of Lazy Deep Learning for Images Recognition in ZZ Photo app
Details of Lazy Deep Learning for Images Recognition in ZZ Photo appDetails of Lazy Deep Learning for Images Recognition in ZZ Photo app
Details of Lazy Deep Learning for Images Recognition in ZZ Photo app
 
Introduction to Deep learning
Introduction to Deep learningIntroduction to Deep learning
Introduction to Deep learning
 
20141003.journal club
20141003.journal club20141003.journal club
20141003.journal club
 
[PR12] Inception and Xception - Jaejun Yoo
[PR12] Inception and Xception - Jaejun Yoo[PR12] Inception and Xception - Jaejun Yoo
[PR12] Inception and Xception - Jaejun Yoo
 
conv_nets.pptx
conv_nets.pptxconv_nets.pptx
conv_nets.pptx
 
(Research Note) Delving deeper into convolutional neural networks for camera ...
(Research Note) Delving deeper into convolutional neural networks for camera ...(Research Note) Delving deeper into convolutional neural networks for camera ...
(Research Note) Delving deeper into convolutional neural networks for camera ...
 
Learning Sparse Neural Networksvia Sensitivity-Driven Regularization
Learning Sparse Neural Networksvia Sensitivity-Driven RegularizationLearning Sparse Neural Networksvia Sensitivity-Driven Regularization
Learning Sparse Neural Networksvia Sensitivity-Driven Regularization
 
Mnist report ppt
Mnist report pptMnist report ppt
Mnist report ppt
 
Modeling perceptual similarity and shift invariance in deep networks
Modeling perceptual similarity and shift invariance in deep networksModeling perceptual similarity and shift invariance in deep networks
Modeling perceptual similarity and shift invariance in deep networks
 
AI&BigData Lab. Артем Чернодуб "Распознавание изображений методом Lazy Deep ...
AI&BigData Lab. Артем Чернодуб  "Распознавание изображений методом Lazy Deep ...AI&BigData Lab. Артем Чернодуб  "Распознавание изображений методом Lazy Deep ...
AI&BigData Lab. Артем Чернодуб "Распознавание изображений методом Lazy Deep ...
 
ct_meeting_final_jcy (1).pdf
ct_meeting_final_jcy (1).pdfct_meeting_final_jcy (1).pdf
ct_meeting_final_jcy (1).pdf
 

Último

Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piececharlottematthew16
 
Search Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfSearch Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfRankYa
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Wonjun Hwang
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfSeasiaInfotech2
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxhariprasad279825
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Enterprise Knowledge
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 3652toLead Limited
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsMemoori
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Commit University
 
Scanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsScanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsRizwan Syed
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embeddingZilliz
 
Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebUiPathCommunity
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostZilliz
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Mattias Andersson
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
DevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenDevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenHervé Boutemy
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyAlfredo García Lavilla
 

Último (20)

Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piece
 
Search Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfSearch Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdf
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdf
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptx
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024
 
DMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special EditionDMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special Edition
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial Buildings
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!
 
Scanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsScanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL Certs
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embedding
 
Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio Web
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
DevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenDevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache Maven
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easy
 

CNN Structure: From LeNet to ShuffleNet

  • 1. CNN Structure: from LeNet to ShuffleNet Dalin Zhang School of CSE, UNSW 17/Jul/2017
  • 2. Preliminary Local receptive fields Shared weights and biases Multiple feature maps Subsampling maps
  • 3. LeNet: Hello World! Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, Gradient-based learning applied to document recognition, Proc. IEEE 86(11): 2278–2324, 1998. C(5x5)-P(2x2) pair repeat Average pooling Sigmoid or tanh activation function
  • 4. ILSVRC ImageNet Large Scale Visual Recognition Challenge (ILSVRC) More than 1.2 Million Images 1000 classes Impressive new CNN structures from ILSVRC www.image- net.org/challenges/LSVRC/
  • 5. AlexNet: ILSVRC 2012 winner C(11x11)P-C(5x5)P-C(3x3)-C(3x3)-C(3x3)P Max pooling Relu activation function 8 layers A. Krizhevsky, I. Sutskever, and G. Hinton, ImageNet Classification with Deep Convolutional Neural Networks, NIPS 2012
  • 6. VGGNet: ILSVRC 2014 2nd All convolutional layer kernels are of size 3x3 MaxPooling of size 2x2 is done after 2 or 3 layers of convolutions Pooling stride is 2 Stacking building blocks of the same shape K. Simonyan and A. Zisserman, Very Deep Convolutional Networks for Large-Scale Image Recognition, ICLR 2015
  • 7. GoogleNet: ILSVRC 2014 Winner Let the network choose the kernel size itself Pointwise convolution (1x1 convolution) reduce parameters 22 layers C. Szegedy et al., Going deeper with convolutions, CVPR 2015
  • 8. ResNet: ILSVRC 2015 Winner Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun, Deep Residual Learning for Image Recognition, CVPR 2016 (Best Paper) ResNet: 152 layers
  • 9. ResNet: ILSVRC 2015 Winner Introduce skip connections Pointwise convolution reduce and restore feature maps 152 layers, top-5 error rate 3.57% vs. 5.1% of human expert Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun, Deep Residual Learning for Image Recognition, CVPR 2016 (Best Paper) Directly performing 3x3 convolutions: Parameters: 256x256x3x3 ~ 600K Residual module structure: Parameters: 64x256x1x1 ~ 16K 64x64x3x3 ~ 36K 256x64x1x1 ~ 16K Total ~70K
  • 10. ResNet: ILSVRC 2015 Winner Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun, Deep Residual Learning for Image Recognition, CVPR 2016 (Best Paper) Problem: with the network depth increasing, accuracy gets saturated (which might be unsurprising) and then degrades rapidly. Deeper network is not easy to optimize. Cause: In some cases some neuron can “die”(output zero) in the training and become ineffective/useless. This can cause information loss, sometimes very important information. Solution: Skip connections carry important information in the previous layer to the next layers.
  • 11. Xception: Depthwise Separable Convolutions François Chollet Xception: Deep Learning with Depthwise Separable Convolution (2017 Apr) Important Hypothesis: The mapping of cross-channels correlations and spatial correlations in the feature maps of convolutional neural networks can be entirely decoupled. output input
  • 12. Xception: Depthwise Separable Convolutions François Chollet Xception: Deep Learning with Depthwise Separable Convolution (2017 Apr)
  • 13. ResNeXt: Group Convolutions ILSVRC 2016 2nd Saining Xie, Ross Girshick, Piotr Dollar, Zhuowen Tu, Kaiming He Aggregated Residual Transformations for Deep Neural Networks (2017 Apr) Introduce group convolution to the ResNet unit, thus introduce a new dimension “cardinality” (the number of groups) to ResNet.
  • 14. ResNeXt: Group Convolutions ILSVRC 2016 2nd More clear case Group convolution reduce the complexity compared to the similar ResNet structure. Gain better performance at the same complexity Saining Xie, Ross Girshick, Piotr Dollar, Zhuowen Tu, Kaiming He Aggregated Residual Transformations for Deep Neural Networks (2017 Apr)
  • 15. ShuffleNeXt: pointwise group conv+channel shuffle Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin and Jian Sun ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices Channel shuffle: help information flow across feature maps (B, g x n, H, W) – reshape(B, g, n, H, W) – transpose(B, n, g, H, W) – reshape(B, g, n, H, W)
  • 16. ShuffleNeXt: pointwise group conv+channel shuffle Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin and Jian Sun ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices Pointwise group convolution: Reduce complexity allowing more feature maps, especially important to small networks
  • 17. Summary  Stack simple structures  Skip connection  Pointwise convolution  Depthwise convolution  Group convolution  Channel shuffle