Module 1 - Introduction to Deep Learning

1. How are Traditional Machine Learning and Deep Learning different when it comes to data needs?
2. How many types of layers do neural networks have? And what are they called?
3. What Deep Learning technique is commonly used for computer vision?

Module 2 - Building Blocks of Deep Learning

1. What are the two main parameters that are learned?
2. Why do we calculate things in a vectorized way?
3. Why do we need a non-linear activation function?
4. What is the goal of backpropagation?
5. What is the role of gradient descent?

Module 4 - Hyperparameters of Neural Networks

Consider the below dataset to answer the first question.

This dataset lists some attributes of different types of red wine. You'd like to use all the features (columns) in the dataset to estimate the quality score of the wine. The quality ranges from 0 to 9.

1. What should be the number of input neurons and output neurons for this project? Shortly explain why.
2. Which activation function should you choose for the output layer for these 3 cases?

(I do not expect you to know all the activation functions by heart. This is a good time to go do some Googling.)

• You want to classify wines into different levels of quality from 0 to 9.
• You want to estimate the income of a household.
• You want to make a model that classifies frogs as male or female based on a photo of them.