Machine Learning: Neural networks from scratch

Focused View

Maxime Vandegar

4:54:46

139 View

1. Introduction

1. Introduction.mp4

02:28

2. Neural networks intuitive explanation.mp4

06:07

2. Forward propagation

1. Forward propagation explanation.mp4

08:05

2. Forward propagation implementation.mp4

10:31

3. Activation function ReLU.mp4

15:16

4. Sequential neural networks.mp4

06:09

3. Image classification

1. Saving and loading neural network parameters.mp4

12:46

2. Image classification part 1.mp4

11:01

3. Image classification part 2.mp4

09:11

4. Activation function Softmax.mp4

04:22

4. Backward propagation

1. Optimization by gradient descent.mp4

10:58

2. Jacobian matrix.mp4

07:38

3. Jacobian matrix implementation.mp4

09:55

4. Chain rule.mp4

04:53

5. Chain rule implementation.mp4

09:59

5. Regression

1. Mean Square Error Loss.mp4

05:51

2. Testing.mp4

14:13

3. Neural network training.mp4

09:26

4. Optimizers.mp4

07:39

5. Regression problem quantitative measure of diabetes progression.mp4

15:22

6. Activation function LogSoftmax.mp4

13:35

7. The Log-Sum-Exp Trick.mp4

15:43

8. Negative Log-Likelihood Loss.mp4

12:15

6. Improvements and tricks

1. Batching Multilayer Perceptron (MLP).mp4

09:19

2. Batching losses.mp4

09:16

3. Batching activation functions.mp4

18:49

4. Jacobian-vector product.mp4

11:32

5. Xavier Initialization.mp4

06:20

7. Image classification and conclusion

1. Image classification.mp4

14:45

2. Conclusion.mp4

01:22

Description

Implementation of neural networks from scratch (Python)

What You'll Learn?

What are neural networks
Implement a neural network from scratch (Python, Java, C, ...)
Training neural networks
Activation functions and the universal approximation theorem
Strengthen your knowledge in Machine Learning and Data Science
Implementation tricks: Jacobian-Vector product & log-sum-exp trick

Who is this for?

For developers who would like to implement a neural network without using dedicated libraries

For those who study machine learning and would like to strengthen their knowledge about neural networks and automatic differentiation frameworks

For those preparing for job interviews in data science

To artificial intelligence enthusiasts

What You Need to Know?

Basic knowledge of programming, algebra and analysis

More details

Description
In this course, we will implement a neural network from scratch, without dedicated libraries. Although we will use the python programming language, at the end of this course, you will be able to implement a neural network in any programming language.

We will see how neural networks work intuitively, and then mathematically. We will also see some important tricks, which allow stabilizing the training of neural networks (log-sum-exp trick), and to prevent the memory used during training from growing exponentially (jacobian-vector product). Without these tricks, most neural networks could not be trained.

We will train our neural networks on real image classification and regression problems. To do so, we will implement different cost functions, as well as several activation functions.

This course is aimed at developers who would like to implement a neural network from scratch as well as those who want to understand how a neural network works from A to Z.

This course is taught using the Python programming language and requires basic programming skills. If you do not have the required background, I recommend that you brush up on your programming skills by taking a crash course in programming. It is also recommended that you have some knowledge of Algebra and Analysis to get the most out of this course.

Concepts covered :
Neural networks
Implementing neural networks from scratch
Gradient descent and Jacobian matrix
The creation of Modules that can be nested in order to create a complex neural architecture
The log-sum-exp trick
Jacobian vector product
Activation functions (ReLU, Softmax, LogSoftmax, ...)
Cost functions (MSELoss, NLLLoss, ...)

This course will be frequently updated, with the addition of bonuses.

Don't wait any longer before launching yourself into the world of machine learning!

Who this course is for:
For developers who would like to implement a neural network without using dedicated libraries
For those who study machine learning and would like to strengthen their knowledge about neural networks and automatic differentiation frameworks
For those preparing for job interviews in data science
To artificial intelligence enthusiasts

User Reviews

Rating

average 0

Total votes0

Focused display

Ingénieur fraîchement diplômé, je suis actuellement chercheur à l'université de Stanford et scientifique collaborateur au CERN. Mes recherches combinent l'intelligence artificielle (principalement le deep learning) et la physique fondamentale.Durant mes études, j'ai été responsable de séances d'exercices dans plusieurs cours universitaires (mécanique des matériaux, électronique numérique, signaux et systèmes,...) et je donne régulièrement des séances de coaching avancées en Python.

Udemy

View courses Udemy

Students take courses primarily to improve job-related skills.Some courses generate credit toward technical certification. Udemy has made a special effort to attract corporate trainers seeking to create coursework for employees of their company.