SlideShare a Scribd company logo
1 of 33
Download to read offline
Convolutional Neural Networks
Sang Jun Lee
Ph.D. candidate, POSTECH
Email: lsj4u0208@postech.ac.kr
EECE695J 전자전기공학특론J(딥러닝기초및철강공정에의활용) – LECTURE 6 (2017. 10. 12)
2
▣ Lecture 4 & 5: Neural Network I & II
1-page Review
Input layer Output layerHidden layer
Perceptron Multilayer perceptron (MLP) Backpropagation
Activation functions
Sigmoid tanh
ReLU Leaky ReLU
Data preprocessing Weight initialization
- RBM (DBN)
- Xavier initialization
𝑾𝑾~𝑵𝑵(𝟎𝟎, 𝝈𝝈𝟐𝟐
)/𝒏𝒏
Optimization
- Gradient descent
- SGD with momentum
- AdaGrad
- Adam
Regularization
- model ensemble
- dropout
CNN components
- Convolutional layer
- Pooling layer
CNN architecture
- General architecture of CNN
- Case study (AlexNet, VGG)
CNN implementation
Other computer vision tasks
- Semantic segmentation
3
Contents
Traditional Method
- Data에 대한 feature vector 설계
- SVM 및 MLP 등의 분류기 사용 및 적은 양의 데이터로 학습 가능
4
Pattern Recognition in Computer Vision
Hand-crafted feature
extractor
Simple trainable
classifier
data output
Feature representation
Deep learning based Method
- Data에 대한 feature representation 및 classifier를 jointly optimize
- End-to-end learning algorithm 설계 가능
5
Pattern Recognition in Computer Vision
Trainable feature
extractor
Trainable classifierdata output
Black box
MLP ( = Fully connected layer)
- 영상 데이터에 대하여 MLP를 이용하여 network를 구성하면..
- 32x32x3 image ≅ 3072-dimensional vector
- 학습 parameter가 증가 → 학습에 필요한 데이터 증가
Convolutional layer
- Convolve a filter with an image (preserve spatial structure)
- Slide over the image spatially and compute dot products
- Parameter sharing & Local connectivity
6
Convolutional layer
MLP의 구성
- Design parameter: Hidden layer 및 각 layer의 neuron 개수
Convolutional layer
- Convolution filter (kernel) size, layer의 개수, 각 layer의 filter 개수 등
7
Convolutional layer
32x32x3 image
5x5x3 filter
height
width
channel
Input layer Output layerHidden layer
Convolutional layer
- Convolve a filter with an image (preserve spatial structure)
- Slide over the image spatially and compute dot products
8
Convolutional layer
32x32x3 image
5x5x3 filter
Convolve (slide) over all
spatial locations
Activation map
(Feature map)
Parameter 수 = 5x5x3 +1(bias)
Consider a second, green filter
9
Convolutional layer
32x32x3 image
5x5x3 filter
Convolve (slide) over all
spatial locations
Activation maps
Depth slice
5x5 filters 6개 사용 → 6 separate activation maps
Activation map의 크기: input size – (kernel size – 1)
10
Convolutional layer
Convolution layer
11
Convolutional layer
Convolutional neural network is a sequence of convolution layers
(interspersed with activation functions)
CONV
ReLU
(6개의
5x5x3 filters)
CONV
ReLU
(10개의
5x5x3 filters)
CONV
ReLU
첫 번째와 두 번째 convolutional layer에 있는 parameter의 수?
12
Convolutional layer
Feature map의 크기
5x5 convolutional filter 사용
Receptive field
- Activation map에서 하나의 값을 결정하는데 사용되는 input에서의 영역 크기
- 5x5 filter로 이루어진 convolutional layer 2개를 통과시킨 activation map의 receptive field = 9x9
12x12
8x8
4x4
CONV
5x5
CONV
5x5
13
Convolutional layer
Zero-padding
- Convolutional layer가 activation map의 크기에 영향을 주지 않도록 하기 위해..
- Pad size = (filter size - 1) / 2
8x8
8x8
Zero-padding: 2
14
Convolutional layer
Stride
- Convolution filter를 얼마만큼 움직일 것인가?
9x9
3x3
5x5 convolution filter
with stride 2
Feature map size = (input size + 2x padding – filter size) +1
15
Pooling layer
Pooling
- Activation map의 크기를 효과적으로 줄이기 위한 down sampling (in spatially)
- max pooling, average pooling, etc.
16
Architecture
General architecture
: Convolution layer (with an activation function) + Pooling layer의 연속
CNN architecture for classification
- Convolutional layer와 pooling layer를 이용하여 spatial domain에서의 feature map size를
1x1로 만드는 것이 핵심
- Output layer의 depth는 classification 하고자 하는 class의 개수와 동일
17
Case study
AlexNet (2012)
8 layers
INPUT – 227x227x3
CONV1 (11x11x3x96 filter with stride 4) – 55x55x96
POOL1 (3x3 pooling with stride 2) – 27x27x96
CONV2 (5x5x96x256 filter with stride 1 and padding 2) – 27x27x256
POOL2 (3x3 pooling with stride 2) – 13x13x256
CONV3 (3x3x256x384 filter with stride 1 and padding 1) – 13x13x384
CONV4 (3x3x384x384 filter with stride 1 and padding 1) – 13x13x384
CONV5 (3x3x384x256 filter with stride 1 and padding 1) – 13x13x384
Fully-connected (4096)
Fully-connected (4096)
Fully-connected (1000)
Feature map size = (input size + 2x padding – filter size) / stride +1
227
227
18
Case study
AlexNet (2012)
ImageNet Large Scale Visual Recognition Challenge (ILSVRC) winners
19
Case study
VGG (2014)
Small filters & Deeper network
왜 3x3의 small filter를 사용하는가?
: 3개의 3x3 convolutional layer (with stride 1)를
이용하여 상대적으로 적은 수의 parameter로
7x7 convolutional layer와 동일한
receptive field의 효과를 낼 수 있음
20
Implementation of CNN
MNIST classification
: 28x28 크기의 숫자 영상 데이터
21
Implementation of CNN
MNIST classification
22
Implementation of CNN
MNIST classification
23
Implementation of CNN
MNIST classification
24
Implementation of CNN
MNIST classification
25
Implementation of CNN
MNIST classification
26
Computer Vision Tasks
Image classification
Other computer vision tasks
Semantic segmentation Localization (single object) Detection (multiple objects) Instance segmentation
27
Semantic segmentation
- Label each pixel in the image with a category label
- Don’t differentiate instances
28
Semantic segmentation
Sliding window based method
29
Semantic segmentation
Fully convolutional manner
(without down- and up-sampling)
30
Semantic segmentation
Unpooling
Learnable upsampling: deconvolution (transposed convolution)
31
Semantic segmentation
Fully convolutional manner
CNN components
- Convolutional layer
- Pooling layer
CNN architecture
- General architecture of CNN
- Case study (AlexNet, VGG)
CNN implementation
Other computer vision tasks
- Semantic segmentation
32
Summary
Deep learning for sequential data
Recurrent neural network
- Vanilla RNN
- LSTM (Long short term memory)
33
Preview (Lecture 7)

More Related Content

What's hot

Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018Universitat Politècnica de Catalunya
 
Optimization for Deep Networks (D2L1 2017 UPC Deep Learning for Computer Vision)
Optimization for Deep Networks (D2L1 2017 UPC Deep Learning for Computer Vision)Optimization for Deep Networks (D2L1 2017 UPC Deep Learning for Computer Vision)
Optimization for Deep Networks (D2L1 2017 UPC Deep Learning for Computer Vision)Universitat Politècnica de Catalunya
 
Applied Deep Learning 11/03 Convolutional Neural Networks
Applied Deep Learning 11/03 Convolutional Neural NetworksApplied Deep Learning 11/03 Convolutional Neural Networks
Applied Deep Learning 11/03 Convolutional Neural NetworksMark Chang
 
Deep Learning for AI (2)
Deep Learning for AI (2)Deep Learning for AI (2)
Deep Learning for AI (2)Dongheon Lee
 
Anomaly detection using deep one class classifier
Anomaly detection using deep one class classifierAnomaly detection using deep one class classifier
Anomaly detection using deep one class classifier홍배 김
 
Deep Neural Networks (D1L2 Insight@DCU Machine Learning Workshop 2017)
Deep Neural Networks (D1L2 Insight@DCU Machine Learning Workshop 2017)Deep Neural Networks (D1L2 Insight@DCU Machine Learning Workshop 2017)
Deep Neural Networks (D1L2 Insight@DCU Machine Learning Workshop 2017)Universitat Politècnica de Catalunya
 
Anomaly Detection and Localization Using GAN and One-Class Classifier
Anomaly Detection and Localization  Using GAN and One-Class ClassifierAnomaly Detection and Localization  Using GAN and One-Class Classifier
Anomaly Detection and Localization Using GAN and One-Class Classifier홍배 김
 
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...Universitat Politècnica de Catalunya
 
The world of loss function
The world of loss functionThe world of loss function
The world of loss function홍배 김
 
Generative Models and Adversarial Training (D2L3 Insight@DCU Machine Learning...
Generative Models and Adversarial Training (D2L3 Insight@DCU Machine Learning...Generative Models and Adversarial Training (D2L3 Insight@DCU Machine Learning...
Generative Models and Adversarial Training (D2L3 Insight@DCU Machine Learning...Universitat Politècnica de Catalunya
 
ARCHITECTURAL CONDITIONING FOR DISENTANGLEMENT OF OBJECT IDENTITY AND POSTURE...
ARCHITECTURAL CONDITIONING FOR DISENTANGLEMENT OF OBJECT IDENTITY AND POSTURE...ARCHITECTURAL CONDITIONING FOR DISENTANGLEMENT OF OBJECT IDENTITY AND POSTURE...
ARCHITECTURAL CONDITIONING FOR DISENTANGLEMENT OF OBJECT IDENTITY AND POSTURE...홍배 김
 
Recent Progress in RNN and NLP
Recent Progress in RNN and NLPRecent Progress in RNN and NLP
Recent Progress in RNN and NLPhytae
 
Convolutional Neural Networks (D1L3 2017 UPC Deep Learning for Computer Vision)
Convolutional Neural Networks (D1L3 2017 UPC Deep Learning for Computer Vision)Convolutional Neural Networks (D1L3 2017 UPC Deep Learning for Computer Vision)
Convolutional Neural Networks (D1L3 2017 UPC Deep Learning for Computer Vision)Universitat Politècnica de Catalunya
 
Convolutional Neural Networks (DLAI D5L1 2017 UPC Deep Learning for Artificia...
Convolutional Neural Networks (DLAI D5L1 2017 UPC Deep Learning for Artificia...Convolutional Neural Networks (DLAI D5L1 2017 UPC Deep Learning for Artificia...
Convolutional Neural Networks (DLAI D5L1 2017 UPC Deep Learning for Artificia...Universitat Politècnica de Catalunya
 
Hands-on Tutorial of Machine Learning in Python
Hands-on Tutorial of Machine Learning in PythonHands-on Tutorial of Machine Learning in Python
Hands-on Tutorial of Machine Learning in PythonChun-Ming Chang
 
Optimization (DLAI D4L1 2017 UPC Deep Learning for Artificial Intelligence)
Optimization (DLAI D4L1 2017 UPC Deep Learning for Artificial Intelligence)Optimization (DLAI D4L1 2017 UPC Deep Learning for Artificial Intelligence)
Optimization (DLAI D4L1 2017 UPC Deep Learning for Artificial Intelligence)Universitat 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
 
Machine learning applications in aerospace domain
Machine learning applications in aerospace domainMachine learning applications in aerospace domain
Machine learning applications in aerospace domain홍배 김
 

What's hot (20)

Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
Convolutional Neural Networks - Veronica Vilaplana - UPC Barcelona 2018
 
Optimization for Deep Networks (D2L1 2017 UPC Deep Learning for Computer Vision)
Optimization for Deep Networks (D2L1 2017 UPC Deep Learning for Computer Vision)Optimization for Deep Networks (D2L1 2017 UPC Deep Learning for Computer Vision)
Optimization for Deep Networks (D2L1 2017 UPC Deep Learning for Computer Vision)
 
Deep Learning for Computer Vision: Deep Networks (UPC 2016)
Deep Learning for Computer Vision: Deep Networks (UPC 2016)Deep Learning for Computer Vision: Deep Networks (UPC 2016)
Deep Learning for Computer Vision: Deep Networks (UPC 2016)
 
Applied Deep Learning 11/03 Convolutional Neural Networks
Applied Deep Learning 11/03 Convolutional Neural NetworksApplied Deep Learning 11/03 Convolutional Neural Networks
Applied Deep Learning 11/03 Convolutional Neural Networks
 
Deep Learning for AI (2)
Deep Learning for AI (2)Deep Learning for AI (2)
Deep Learning for AI (2)
 
Anomaly detection using deep one class classifier
Anomaly detection using deep one class classifierAnomaly detection using deep one class classifier
Anomaly detection using deep one class classifier
 
Deep Neural Networks (D1L2 Insight@DCU Machine Learning Workshop 2017)
Deep Neural Networks (D1L2 Insight@DCU Machine Learning Workshop 2017)Deep Neural Networks (D1L2 Insight@DCU Machine Learning Workshop 2017)
Deep Neural Networks (D1L2 Insight@DCU Machine Learning Workshop 2017)
 
Anomaly Detection and Localization Using GAN and One-Class Classifier
Anomaly Detection and Localization  Using GAN and One-Class ClassifierAnomaly Detection and Localization  Using GAN and One-Class Classifier
Anomaly Detection and Localization Using GAN and One-Class Classifier
 
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
 
The world of loss function
The world of loss functionThe world of loss function
The world of loss function
 
Generative Models and Adversarial Training (D2L3 Insight@DCU Machine Learning...
Generative Models and Adversarial Training (D2L3 Insight@DCU Machine Learning...Generative Models and Adversarial Training (D2L3 Insight@DCU Machine Learning...
Generative Models and Adversarial Training (D2L3 Insight@DCU Machine Learning...
 
ARCHITECTURAL CONDITIONING FOR DISENTANGLEMENT OF OBJECT IDENTITY AND POSTURE...
ARCHITECTURAL CONDITIONING FOR DISENTANGLEMENT OF OBJECT IDENTITY AND POSTURE...ARCHITECTURAL CONDITIONING FOR DISENTANGLEMENT OF OBJECT IDENTITY AND POSTURE...
ARCHITECTURAL CONDITIONING FOR DISENTANGLEMENT OF OBJECT IDENTITY AND POSTURE...
 
Deep Learning for Computer Vision: Visualization (UPC 2016)
Deep Learning for Computer Vision: Visualization (UPC 2016)Deep Learning for Computer Vision: Visualization (UPC 2016)
Deep Learning for Computer Vision: Visualization (UPC 2016)
 
Recent Progress in RNN and NLP
Recent Progress in RNN and NLPRecent Progress in RNN and NLP
Recent Progress in RNN and NLP
 
Convolutional Neural Networks (D1L3 2017 UPC Deep Learning for Computer Vision)
Convolutional Neural Networks (D1L3 2017 UPC Deep Learning for Computer Vision)Convolutional Neural Networks (D1L3 2017 UPC Deep Learning for Computer Vision)
Convolutional Neural Networks (D1L3 2017 UPC Deep Learning for Computer Vision)
 
Convolutional Neural Networks (DLAI D5L1 2017 UPC Deep Learning for Artificia...
Convolutional Neural Networks (DLAI D5L1 2017 UPC Deep Learning for Artificia...Convolutional Neural Networks (DLAI D5L1 2017 UPC Deep Learning for Artificia...
Convolutional Neural Networks (DLAI D5L1 2017 UPC Deep Learning for Artificia...
 
Hands-on Tutorial of Machine Learning in Python
Hands-on Tutorial of Machine Learning in PythonHands-on Tutorial of Machine Learning in Python
Hands-on Tutorial of Machine Learning in Python
 
Optimization (DLAI D4L1 2017 UPC Deep Learning for Artificial Intelligence)
Optimization (DLAI D4L1 2017 UPC Deep Learning for Artificial Intelligence)Optimization (DLAI D4L1 2017 UPC Deep Learning for Artificial Intelligence)
Optimization (DLAI D4L1 2017 UPC Deep Learning for Artificial Intelligence)
 
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)
 
Machine learning applications in aerospace domain
Machine learning applications in aerospace domainMachine learning applications in aerospace domain
Machine learning applications in aerospace domain
 

Similar to Lecture 6: Convolutional Neural Networks

ImageNet classification with deep convolutional neural networks(2012)
ImageNet classification with deep convolutional neural networks(2012)ImageNet classification with deep convolutional neural networks(2012)
ImageNet classification with deep convolutional neural networks(2012)WoochulShin10
 
Lecture 5: Convolutional Neural Network Models
Lecture 5: Convolutional Neural Network ModelsLecture 5: Convolutional Neural Network Models
Lecture 5: Convolutional Neural Network ModelsMohamed Loey
 
Convolutional Neural Networks
Convolutional Neural NetworksConvolutional Neural Networks
Convolutional Neural Networksmilad abbasi
 
Trackster Pruning at the CMS High-Granularity Calorimeter
Trackster Pruning at the CMS High-Granularity CalorimeterTrackster Pruning at the CMS High-Granularity Calorimeter
Trackster Pruning at the CMS High-Granularity CalorimeterYousef Fadila
 
Machine Learning - Convolutional Neural Network
Machine Learning - Convolutional Neural NetworkMachine Learning - Convolutional Neural Network
Machine Learning - Convolutional Neural NetworkRichard Kuo
 
Hardware Acceleration for Machine Learning
Hardware Acceleration for Machine LearningHardware Acceleration for Machine Learning
Hardware Acceleration for Machine LearningCastLabKAIST
 
Efficient de cvpr_2020_paper
Efficient de cvpr_2020_paperEfficient de cvpr_2020_paper
Efficient de cvpr_2020_papershanullah3
 
Multilayer Perceptron (DLAI D1L2 2017 UPC Deep Learning for Artificial Intell...
Multilayer Perceptron (DLAI D1L2 2017 UPC Deep Learning for Artificial Intell...Multilayer Perceptron (DLAI D1L2 2017 UPC Deep Learning for Artificial Intell...
Multilayer Perceptron (DLAI D1L2 2017 UPC Deep Learning for Artificial Intell...Universitat Politècnica de Catalunya
 
Cvpr 2018 papers review (efficient computing)
Cvpr 2018 papers review (efficient computing)Cvpr 2018 papers review (efficient computing)
Cvpr 2018 papers review (efficient computing)DonghyunKang12
 
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 Network Models - Deep Learning
Convolutional Neural Network Models - Deep LearningConvolutional Neural Network Models - Deep Learning
Convolutional Neural Network Models - Deep LearningMohamed Loey
 
Restricting the Flow: Information Bottlenecks for Attribution
Restricting the Flow: Information Bottlenecks for AttributionRestricting the Flow: Information Bottlenecks for Attribution
Restricting the Flow: Information Bottlenecks for Attributiontaeseon ryu
 
AI optimizing HPC simulations (presentation from 6th EULAG Workshop)
AI optimizing HPC simulations (presentation from  6th EULAG Workshop)AI optimizing HPC simulations (presentation from  6th EULAG Workshop)
AI optimizing HPC simulations (presentation from 6th EULAG Workshop)byteLAKE
 
Convolutional Neural Networks (CNN)
Convolutional Neural Networks (CNN)Convolutional Neural Networks (CNN)
Convolutional Neural Networks (CNN)Gaurav Mittal
 
Implementation of the fully adaptive radar framework: Practical limitations
Implementation of the fully adaptive radar framework: Practical limitationsImplementation of the fully adaptive radar framework: Practical limitations
Implementation of the fully adaptive radar framework: Practical limitationsLuis Úbeda Medina
 

Similar to Lecture 6: Convolutional Neural Networks (20)

ImageNet classification with deep convolutional neural networks(2012)
ImageNet classification with deep convolutional neural networks(2012)ImageNet classification with deep convolutional neural networks(2012)
ImageNet classification with deep convolutional neural networks(2012)
 
Lecture 5: Convolutional Neural Network Models
Lecture 5: Convolutional Neural Network ModelsLecture 5: Convolutional Neural Network Models
Lecture 5: Convolutional Neural Network Models
 
B.tech_project_ppt.pptx
B.tech_project_ppt.pptxB.tech_project_ppt.pptx
B.tech_project_ppt.pptx
 
Cnn
CnnCnn
Cnn
 
Convolutional Neural Networks
Convolutional Neural NetworksConvolutional Neural Networks
Convolutional Neural Networks
 
Trackster Pruning at the CMS High-Granularity Calorimeter
Trackster Pruning at the CMS High-Granularity CalorimeterTrackster Pruning at the CMS High-Granularity Calorimeter
Trackster Pruning at the CMS High-Granularity Calorimeter
 
Machine Learning - Convolutional Neural Network
Machine Learning - Convolutional Neural NetworkMachine Learning - Convolutional Neural Network
Machine Learning - Convolutional Neural Network
 
Hardware Acceleration for Machine Learning
Hardware Acceleration for Machine LearningHardware Acceleration for Machine Learning
Hardware Acceleration for Machine Learning
 
Efficient de cvpr_2020_paper
Efficient de cvpr_2020_paperEfficient de cvpr_2020_paper
Efficient de cvpr_2020_paper
 
28 01-2021-05
28 01-2021-0528 01-2021-05
28 01-2021-05
 
Multilayer Perceptron (DLAI D1L2 2017 UPC Deep Learning for Artificial Intell...
Multilayer Perceptron (DLAI D1L2 2017 UPC Deep Learning for Artificial Intell...Multilayer Perceptron (DLAI D1L2 2017 UPC Deep Learning for Artificial Intell...
Multilayer Perceptron (DLAI D1L2 2017 UPC Deep Learning for Artificial Intell...
 
Cvpr 2018 papers review (efficient computing)
Cvpr 2018 papers review (efficient computing)Cvpr 2018 papers review (efficient computing)
Cvpr 2018 papers review (efficient computing)
 
Conditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN DecodersConditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN Decoders
 
Deep Learning
Deep LearningDeep Learning
Deep Learning
 
Convolutional Neural Network Models - Deep Learning
Convolutional Neural Network Models - Deep LearningConvolutional Neural Network Models - Deep Learning
Convolutional Neural Network Models - Deep Learning
 
Restricting the Flow: Information Bottlenecks for Attribution
Restricting the Flow: Information Bottlenecks for AttributionRestricting the Flow: Information Bottlenecks for Attribution
Restricting the Flow: Information Bottlenecks for Attribution
 
AI optimizing HPC simulations (presentation from 6th EULAG Workshop)
AI optimizing HPC simulations (presentation from  6th EULAG Workshop)AI optimizing HPC simulations (presentation from  6th EULAG Workshop)
AI optimizing HPC simulations (presentation from 6th EULAG Workshop)
 
Convolutional Neural Networks (CNN)
Convolutional Neural Networks (CNN)Convolutional Neural Networks (CNN)
Convolutional Neural Networks (CNN)
 
Implementation of the fully adaptive radar framework: Practical limitations
Implementation of the fully adaptive radar framework: Practical limitationsImplementation of the fully adaptive radar framework: Practical limitations
Implementation of the fully adaptive radar framework: Practical limitations
 
Multilayer Perceptron - Elisa Sayrol - UPC Barcelona 2018
Multilayer Perceptron - Elisa Sayrol - UPC Barcelona 2018Multilayer Perceptron - Elisa Sayrol - UPC Barcelona 2018
Multilayer Perceptron - Elisa Sayrol - UPC Barcelona 2018
 

More from Sang Jun Lee

[5분 논문요약] Structured Knowledge Distillation for Semantic Segmentation
[5분 논문요약] Structured Knowledge Distillation for Semantic Segmentation[5분 논문요약] Structured Knowledge Distillation for Semantic Segmentation
[5분 논문요약] Structured Knowledge Distillation for Semantic SegmentationSang Jun Lee
 
Lecture 7: Recurrent Neural Networks
Lecture 7: Recurrent Neural NetworksLecture 7: Recurrent Neural Networks
Lecture 7: Recurrent Neural NetworksSang Jun Lee
 
Lecture 4: Neural Networks I
Lecture 4: Neural Networks ILecture 4: Neural Networks I
Lecture 4: Neural Networks ISang Jun Lee
 
Lecture 3: Unsupervised Learning
Lecture 3: Unsupervised LearningLecture 3: Unsupervised Learning
Lecture 3: Unsupervised LearningSang Jun Lee
 
Lecture 2: Supervised Learning
Lecture 2: Supervised LearningLecture 2: Supervised Learning
Lecture 2: Supervised LearningSang Jun Lee
 
Lecture 1: Introduction to Python and TensorFlow
Lecture 1: Introduction to Python and TensorFlowLecture 1: Introduction to Python and TensorFlow
Lecture 1: Introduction to Python and TensorFlowSang Jun Lee
 

More from Sang Jun Lee (6)

[5분 논문요약] Structured Knowledge Distillation for Semantic Segmentation
[5분 논문요약] Structured Knowledge Distillation for Semantic Segmentation[5분 논문요약] Structured Knowledge Distillation for Semantic Segmentation
[5분 논문요약] Structured Knowledge Distillation for Semantic Segmentation
 
Lecture 7: Recurrent Neural Networks
Lecture 7: Recurrent Neural NetworksLecture 7: Recurrent Neural Networks
Lecture 7: Recurrent Neural Networks
 
Lecture 4: Neural Networks I
Lecture 4: Neural Networks ILecture 4: Neural Networks I
Lecture 4: Neural Networks I
 
Lecture 3: Unsupervised Learning
Lecture 3: Unsupervised LearningLecture 3: Unsupervised Learning
Lecture 3: Unsupervised Learning
 
Lecture 2: Supervised Learning
Lecture 2: Supervised LearningLecture 2: Supervised Learning
Lecture 2: Supervised Learning
 
Lecture 1: Introduction to Python and TensorFlow
Lecture 1: Introduction to Python and TensorFlowLecture 1: Introduction to Python and TensorFlow
Lecture 1: Introduction to Python and TensorFlow
 

Recently uploaded

Software and Systems Engineering Standards: Verification and Validation of Sy...
Software and Systems Engineering Standards: Verification and Validation of Sy...Software and Systems Engineering Standards: Verification and Validation of Sy...
Software and Systems Engineering Standards: Verification and Validation of Sy...VICTOR MAESTRE RAMIREZ
 
power system scada applications and uses
power system scada applications and usespower system scada applications and uses
power system scada applications and usesDevarapalliHaritha
 
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICSAPPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICSKurinjimalarL3
 
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...ZTE
 
Past, Present and Future of Generative AI
Past, Present and Future of Generative AIPast, Present and Future of Generative AI
Past, Present and Future of Generative AIabhishek36461
 
Oxy acetylene welding presentation note.
Oxy acetylene welding presentation note.Oxy acetylene welding presentation note.
Oxy acetylene welding presentation note.eptoze12
 
Biology for Computer Engineers Course Handout.pptx
Biology for Computer Engineers Course Handout.pptxBiology for Computer Engineers Course Handout.pptx
Biology for Computer Engineers Course Handout.pptxDeepakSakkari2
 
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escortsranjana rawat
 
Heart Disease Prediction using machine learning.pptx
Heart Disease Prediction using machine learning.pptxHeart Disease Prediction using machine learning.pptx
Heart Disease Prediction using machine learning.pptxPoojaBan
 
Internship report on mechanical engineering
Internship report on mechanical engineeringInternship report on mechanical engineering
Internship report on mechanical engineeringmalavadedarshan25
 
Application of Residue Theorem to evaluate real integrations.pptx
Application of Residue Theorem to evaluate real integrations.pptxApplication of Residue Theorem to evaluate real integrations.pptx
Application of Residue Theorem to evaluate real integrations.pptx959SahilShah
 
Current Transformer Drawing and GTP for MSETCL
Current Transformer Drawing and GTP for MSETCLCurrent Transformer Drawing and GTP for MSETCL
Current Transformer Drawing and GTP for MSETCLDeelipZope
 
Architect Hassan Khalil Portfolio for 2024
Architect Hassan Khalil Portfolio for 2024Architect Hassan Khalil Portfolio for 2024
Architect Hassan Khalil Portfolio for 2024hassan khalil
 
Call Girls Narol 7397865700 Independent Call Girls
Call Girls Narol 7397865700 Independent Call GirlsCall Girls Narol 7397865700 Independent Call Girls
Call Girls Narol 7397865700 Independent Call Girlsssuser7cb4ff
 
HARMONY IN THE HUMAN BEING - Unit-II UHV-2
HARMONY IN THE HUMAN BEING - Unit-II UHV-2HARMONY IN THE HUMAN BEING - Unit-II UHV-2
HARMONY IN THE HUMAN BEING - Unit-II UHV-2RajaP95
 
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Dr.Costas Sachpazis
 
Call Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile serviceCall Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile servicerehmti665
 

Recently uploaded (20)

Exploring_Network_Security_with_JA3_by_Rakesh Seal.pptx
Exploring_Network_Security_with_JA3_by_Rakesh Seal.pptxExploring_Network_Security_with_JA3_by_Rakesh Seal.pptx
Exploring_Network_Security_with_JA3_by_Rakesh Seal.pptx
 
★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR
★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR
★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR
 
Software and Systems Engineering Standards: Verification and Validation of Sy...
Software and Systems Engineering Standards: Verification and Validation of Sy...Software and Systems Engineering Standards: Verification and Validation of Sy...
Software and Systems Engineering Standards: Verification and Validation of Sy...
 
power system scada applications and uses
power system scada applications and usespower system scada applications and uses
power system scada applications and uses
 
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICSAPPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
 
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
 
Past, Present and Future of Generative AI
Past, Present and Future of Generative AIPast, Present and Future of Generative AI
Past, Present and Future of Generative AI
 
Oxy acetylene welding presentation note.
Oxy acetylene welding presentation note.Oxy acetylene welding presentation note.
Oxy acetylene welding presentation note.
 
Biology for Computer Engineers Course Handout.pptx
Biology for Computer Engineers Course Handout.pptxBiology for Computer Engineers Course Handout.pptx
Biology for Computer Engineers Course Handout.pptx
 
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
 
Heart Disease Prediction using machine learning.pptx
Heart Disease Prediction using machine learning.pptxHeart Disease Prediction using machine learning.pptx
Heart Disease Prediction using machine learning.pptx
 
Internship report on mechanical engineering
Internship report on mechanical engineeringInternship report on mechanical engineering
Internship report on mechanical engineering
 
Call Us -/9953056974- Call Girls In Vikaspuri-/- Delhi NCR
Call Us -/9953056974- Call Girls In Vikaspuri-/- Delhi NCRCall Us -/9953056974- Call Girls In Vikaspuri-/- Delhi NCR
Call Us -/9953056974- Call Girls In Vikaspuri-/- Delhi NCR
 
Application of Residue Theorem to evaluate real integrations.pptx
Application of Residue Theorem to evaluate real integrations.pptxApplication of Residue Theorem to evaluate real integrations.pptx
Application of Residue Theorem to evaluate real integrations.pptx
 
Current Transformer Drawing and GTP for MSETCL
Current Transformer Drawing and GTP for MSETCLCurrent Transformer Drawing and GTP for MSETCL
Current Transformer Drawing and GTP for MSETCL
 
Architect Hassan Khalil Portfolio for 2024
Architect Hassan Khalil Portfolio for 2024Architect Hassan Khalil Portfolio for 2024
Architect Hassan Khalil Portfolio for 2024
 
Call Girls Narol 7397865700 Independent Call Girls
Call Girls Narol 7397865700 Independent Call GirlsCall Girls Narol 7397865700 Independent Call Girls
Call Girls Narol 7397865700 Independent Call Girls
 
HARMONY IN THE HUMAN BEING - Unit-II UHV-2
HARMONY IN THE HUMAN BEING - Unit-II UHV-2HARMONY IN THE HUMAN BEING - Unit-II UHV-2
HARMONY IN THE HUMAN BEING - Unit-II UHV-2
 
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
 
Call Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile serviceCall Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile service
 

Lecture 6: Convolutional Neural Networks

  • 1. Convolutional Neural Networks Sang Jun Lee Ph.D. candidate, POSTECH Email: lsj4u0208@postech.ac.kr EECE695J 전자전기공학특론J(딥러닝기초및철강공정에의활용) – LECTURE 6 (2017. 10. 12)
  • 2. 2 ▣ Lecture 4 & 5: Neural Network I & II 1-page Review Input layer Output layerHidden layer Perceptron Multilayer perceptron (MLP) Backpropagation Activation functions Sigmoid tanh ReLU Leaky ReLU Data preprocessing Weight initialization - RBM (DBN) - Xavier initialization 𝑾𝑾~𝑵𝑵(𝟎𝟎, 𝝈𝝈𝟐𝟐 )/𝒏𝒏 Optimization - Gradient descent - SGD with momentum - AdaGrad - Adam Regularization - model ensemble - dropout
  • 3. CNN components - Convolutional layer - Pooling layer CNN architecture - General architecture of CNN - Case study (AlexNet, VGG) CNN implementation Other computer vision tasks - Semantic segmentation 3 Contents
  • 4. Traditional Method - Data에 대한 feature vector 설계 - SVM 및 MLP 등의 분류기 사용 및 적은 양의 데이터로 학습 가능 4 Pattern Recognition in Computer Vision Hand-crafted feature extractor Simple trainable classifier data output Feature representation
  • 5. Deep learning based Method - Data에 대한 feature representation 및 classifier를 jointly optimize - End-to-end learning algorithm 설계 가능 5 Pattern Recognition in Computer Vision Trainable feature extractor Trainable classifierdata output Black box
  • 6. MLP ( = Fully connected layer) - 영상 데이터에 대하여 MLP를 이용하여 network를 구성하면.. - 32x32x3 image ≅ 3072-dimensional vector - 학습 parameter가 증가 → 학습에 필요한 데이터 증가 Convolutional layer - Convolve a filter with an image (preserve spatial structure) - Slide over the image spatially and compute dot products - Parameter sharing & Local connectivity 6 Convolutional layer
  • 7. MLP의 구성 - Design parameter: Hidden layer 및 각 layer의 neuron 개수 Convolutional layer - Convolution filter (kernel) size, layer의 개수, 각 layer의 filter 개수 등 7 Convolutional layer 32x32x3 image 5x5x3 filter height width channel Input layer Output layerHidden layer
  • 8. Convolutional layer - Convolve a filter with an image (preserve spatial structure) - Slide over the image spatially and compute dot products 8 Convolutional layer 32x32x3 image 5x5x3 filter Convolve (slide) over all spatial locations Activation map (Feature map) Parameter 수 = 5x5x3 +1(bias)
  • 9. Consider a second, green filter 9 Convolutional layer 32x32x3 image 5x5x3 filter Convolve (slide) over all spatial locations Activation maps Depth slice
  • 10. 5x5 filters 6개 사용 → 6 separate activation maps Activation map의 크기: input size – (kernel size – 1) 10 Convolutional layer Convolution layer
  • 11. 11 Convolutional layer Convolutional neural network is a sequence of convolution layers (interspersed with activation functions) CONV ReLU (6개의 5x5x3 filters) CONV ReLU (10개의 5x5x3 filters) CONV ReLU 첫 번째와 두 번째 convolutional layer에 있는 parameter의 수?
  • 12. 12 Convolutional layer Feature map의 크기 5x5 convolutional filter 사용 Receptive field - Activation map에서 하나의 값을 결정하는데 사용되는 input에서의 영역 크기 - 5x5 filter로 이루어진 convolutional layer 2개를 통과시킨 activation map의 receptive field = 9x9 12x12 8x8 4x4 CONV 5x5 CONV 5x5
  • 13. 13 Convolutional layer Zero-padding - Convolutional layer가 activation map의 크기에 영향을 주지 않도록 하기 위해.. - Pad size = (filter size - 1) / 2 8x8 8x8 Zero-padding: 2
  • 14. 14 Convolutional layer Stride - Convolution filter를 얼마만큼 움직일 것인가? 9x9 3x3 5x5 convolution filter with stride 2 Feature map size = (input size + 2x padding – filter size) +1
  • 15. 15 Pooling layer Pooling - Activation map의 크기를 효과적으로 줄이기 위한 down sampling (in spatially) - max pooling, average pooling, etc.
  • 16. 16 Architecture General architecture : Convolution layer (with an activation function) + Pooling layer의 연속 CNN architecture for classification - Convolutional layer와 pooling layer를 이용하여 spatial domain에서의 feature map size를 1x1로 만드는 것이 핵심 - Output layer의 depth는 classification 하고자 하는 class의 개수와 동일
  • 17. 17 Case study AlexNet (2012) 8 layers INPUT – 227x227x3 CONV1 (11x11x3x96 filter with stride 4) – 55x55x96 POOL1 (3x3 pooling with stride 2) – 27x27x96 CONV2 (5x5x96x256 filter with stride 1 and padding 2) – 27x27x256 POOL2 (3x3 pooling with stride 2) – 13x13x256 CONV3 (3x3x256x384 filter with stride 1 and padding 1) – 13x13x384 CONV4 (3x3x384x384 filter with stride 1 and padding 1) – 13x13x384 CONV5 (3x3x384x256 filter with stride 1 and padding 1) – 13x13x384 Fully-connected (4096) Fully-connected (4096) Fully-connected (1000) Feature map size = (input size + 2x padding – filter size) / stride +1 227 227
  • 18. 18 Case study AlexNet (2012) ImageNet Large Scale Visual Recognition Challenge (ILSVRC) winners
  • 19. 19 Case study VGG (2014) Small filters & Deeper network 왜 3x3의 small filter를 사용하는가? : 3개의 3x3 convolutional layer (with stride 1)를 이용하여 상대적으로 적은 수의 parameter로 7x7 convolutional layer와 동일한 receptive field의 효과를 낼 수 있음
  • 20. 20 Implementation of CNN MNIST classification : 28x28 크기의 숫자 영상 데이터
  • 26. 26 Computer Vision Tasks Image classification Other computer vision tasks Semantic segmentation Localization (single object) Detection (multiple objects) Instance segmentation
  • 27. 27 Semantic segmentation - Label each pixel in the image with a category label - Don’t differentiate instances
  • 29. 29 Semantic segmentation Fully convolutional manner (without down- and up-sampling)
  • 30. 30 Semantic segmentation Unpooling Learnable upsampling: deconvolution (transposed convolution)
  • 32. CNN components - Convolutional layer - Pooling layer CNN architecture - General architecture of CNN - Case study (AlexNet, VGG) CNN implementation Other computer vision tasks - Semantic segmentation 32 Summary
  • 33. Deep learning for sequential data Recurrent neural network - Vanilla RNN - LSTM (Long short term memory) 33 Preview (Lecture 7)