Introduction

Question 1

What is the difference between AI and machine learning?

How do they work? What are they based on?

Answer 1

AI is to assist human in probleme solving using formal rules
Machine learning learns from data, there are no hardcored rules. It defines relevant features and map them to the desired output.

The choice of those features is crucial to the performance of the model

Question 2

Cite a few examples of variations in data that might make defininf relevant features difficult

The questions is related to images

Answer 2

• Different scales
• number of instances
• Image cluter
• deformability
• amount of texture
• color
• shape
• real world size.

Example, a dog can be a chihuaha, a dalmatian, or a german shepherd

Question 3

What is representation learning? What is deep learning?

Hint: Deep learning is achieving represnetation learning. How?

Answer 3

Representation learning=learning optimal features from the data.
Deep learning =does this gradually by extracting firstly easy and then harder features. It learns complex features going through the easy ones first.

Question 4

What is a neuron, a layer and a neural network?

Answer 4

Neuron=function that computes features.
Layers= neurons arranged together.
Neural network=set of all interconnected neurons

The goal is to find a pattern locally first using neurons

Other definition for neuron: multiple-input single-output parametric nonlinear function