​

 

 

Course Overview

​

This course will help candidates to gain understanding of the underlying concepts of Deep Learning.  Students will learn about supervised and unsupervised Deep learning networks and will be given in depth understanding of the intuition behind Artificial Neural Networks (ANNs), Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs) and Self Organising Maps (SOMs).  The candidates will also implement real-life-use cases of the above types of networks in Python.

 

 

Learning Outcome

​

On successful completion of the course, a candidate will be able to:

 

1. Choose the type of deep learning networks for solving a business problem at hand.

2. Prepare data and build an appropriate neural network in Python.

3. Train the neural network with data for optimising loss function.

4. Test accuracy of model and do predictions on future data using the predictive model.

 

Course Duration

36 Hours

 

                                                  

                           Course Curriculum

 

 

 

1.  Introduction to Deep Learning

​

• What is Deep Learning?

• How is Deep Learning different from Machine Learning?

• Advantages of Deep Learning over Machine Learning.

• Supervised and Unsupervised Deep Learning

 

 

2.  Artificial Neural Networks (ANN)

​

• The Neuron

• How do Neural Networks work?

• The Activation Function

• How do Neural Networks Learn?

• Gradient Descent

• Stochastic Gradient Descent

• Backpropagation

• Epochs

• Building and ANN

 

 

3.  Convolutional Neural Networks (CNN)

​

• What are Convolutional Neural Networks?

• Convolution Operation

• ReLU layer

• Pooling Flattening

• Full Connection with ANN

• Softmax and Cross-Entropy

• Building a CNN

 

 

4.  Recurrent Neural Networks (RNN)

​

• The idea behind Recurrent Neural Networks

• The Vanishing Gradient Problem

• LSTMs

• Practical Intuition

• Building a RNN

• Evaluating the RNN

• Improving the RNN

 

 

5.  Self-Organising Maps (SOM)

​

• How do Self-Organising Maps work?

• K-Means clustering (refresher)

• How do Self-Organising Maps learn?

• Live SOM example

• Reading an advanced SOM

• Building a SOM