Scribd
Upload
2 views19 pages

Deep Learning Introduction

Deep Learning is a subfield of Machine Learning that utilizes Artificial Neural Networks to mimic human decision-making processes. It involves training models to recognize patterns, such as handwritten digits, through optimization techniques like Gradient Descent. The document outlines the components of neural networks, training data preparation, cost functions, and the training process including forward pass and error backpropagation.

Uploaded by

elamranihoussameddine404
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views19 pages

Deep Learning Introduction

Deep Learning is a subfield of Machine Learning that utilizes Artificial Neural Networks to mimic human decision-making processes. It involves training models to recognize patterns, such as handwritten digits, through optimization techniques like Gradient Descent. The document outlines the components of neural networks, training data preparation, cost functions, and the training process including forward pass and error backpropagation.

Uploaded by

elamranihoussameddine404
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Table of Contents

3 Deep Learning

▶ Motivation

▶ Traditional Machine Learning

▶ Deep Learning

▶ Scientific Machine Learning

▶ Scientific Machine Learning : Methods and Applications

▶ Summary
Deep Learning
What is Deep Learning?

• Deep Learning (DL) is the


subfield of Machine Learning
which uses an ”Artificial Neural
Network” (ANN) (A simulation of
a human’s neurons network) to
make decisions just like our brain
makes decisions using neurons.
• Deep Learning tries to mimic the
way the human brain operates.
• As we learn from our mistakes, a
Deep Learning model also learns
from its previous decisions.
Deep Learning
Deep Learning vs. Machine Learning: What’s the difference?
Deep Learning
Handwriting Digit Recognition Example

• Handwritten digit
recognition is the process to
provide the ability to machines to
recognize human handwritten
digits.
• It is not an easy task for
the machine because handwritten
digits are not perfect, vary from
person-to-person, and can be
made with many different flavors.
Deep Learning
Handwriting Digit Recognition Example
Deep Learning
Handwriting Digit Recognition Example
Deep Learning
Element of Neural Network : Artificial Neuron (Perceptron)
Deep Learning
Neural Network (NN)
Deep Learning
Example of Neural Network
Deep Learning
Example of Neural Network
Deep Learning
Example of Neural Network
Deep Learning
Matrix Operation
Deep Learning
How to set network parameters (Wi , bi ) i=1,...,L ?

Set the Network parameters θ such that:


• Input = ”2”: y2 has the maximum value. How to let the neural
network achieve this?
• Input , ”2”: yi,2 has the minimum value.
Deep Learning
Training Data

Preparing training data: images and their labels.

Using the training


 1 data to2 find
 the network
1 2
parameters θ = W , b , W , b , . . . , W L, b L :

Optimization Problem !
Deep Learning
Cost Function L(θ)

Given a set of network parameters θ, each example/input k has a


Cost Value Lk (θ).

The Cost function Lk (θ) can be Euclidean distance or cross


entropy of the Neural Network Output and Target.
Deep Learning
Total Cost/Loss Function
For all training data ...
• Total Cost/Loss :
L(θ) = R k
Í
k=1 L (θ).
• How bad the network
parameters θ is on this
task.

Find the Neural Network


Optimal Parameters θ⋆ that
minimize the Total
Cost/Loss L(θ) value:
Training Process.

Gradient Descent
Algorithm
Deep Learning
Gradient Descent Algorithm & Updating Weight : Training Process
Deep Learning
Training Process : Forward pass and Error Backpropagation
(Expectation-Maximization Algorithm)
Deep Learning
Popular Tools

You might also like