View markdown source on GitHub

Convolutional neural networks (CNN) Deep Learning - Part 3

Contributors

Kaivan Kamali

Questions

Objectives

Requirements

last_modification Last modification: Jul 9, 2021

What is a convolutional neural network (CNN)?

Speaker Notes

What is a convolutional neural network (CNN)?


Convolutional Neural Network (CNN)


Feedforward neural networks (FNN)

Neurons forming the input, output, and hidden layers of a multi-layer feedforward neural network


Limitations of FNN


Limitations of FNN


Limitations of FNN


Inspiration for CNN


Inspiration for CNN


Architecture of CNN


Input layer


Convolution layer


3 by 3 Filter

A 3 by 3 filter applied to a 4 by 4 image, resulting in a 2 by 2 image


Covolution operator parameters


Filter size


Padding


3 by 3 filter with padding of 1

A 3 by 3 filter applied to a 5 by 5 image, with padding of 1, resulting in a 5 by 5 image


Stride


3 by 3 filter with stride of 2

A 3 by 3 filter applied to a 5 by 5 image, with stride of 2, resulting in a 2 by 2 image


Dilation


3 by 3 filter with dilation of 2

A 3 by 3 filter applied to a 7 by 7 image, with dilation of 2, resulting in a 3 by 3 image


Activation function


Relu activation function

Two matrices representing filter output before and after ReLU activation function is applied


Single channel 2D convolution

One matrix representing an input vector and another matrix representing a filter, along with calculation for single input channel two dimensional convolution operation


Triple channel 2D convolution

Three matrices representing an input vector and another three matrices representing a filter, along with calculation for multiple input channel two dimensional convolution operation


Triple channel 2D convolution in 3D

Multiple cubes representing input vector, filter, and output in a 3 channel 2 dimensional convolution operation


Change channel size


Pooling layer


Fully connected layer


An example CNN

A convolutional neural network with 3 convolution layers followed by 3 pooling layers


An example CNN


MNIST dataset

Classification of MNIST images with CNN


For references, please see tutorial’s References section


Screenshot of the gtn stats page with 21 topics, 170 tutorials, 159 contributors, 16 scientific topics, and a growing community

Speaker Notes


Getting Help

Speaker Notes


Join an event

Event schedule

Speaker Notes


Thank you!

This material is the result of a collaborative work. Thanks to the Galaxy Training Network and all the contributors! Galaxy Training Network This material is licensed under the Creative Commons Attribution 4.0 International License.