1) Introduction and Motivation

Question 1

What is Supervised Learning

Answer 1

The goal is to learn a function that maps inputs x1,…,xN to labels y1,…,yN, so that it can predict the label y for new, unseen inputs x

Question 2

What are examples of supervised learning problems

Answer 2

• Simple curve fitting (linear regression)
• Image segmentation

Question 3

What is Unsupervised Learning

Answer 3

In unsupervised learning, we are only given inputs x1, . . . , xN but no corresponding labels. The goal is to sort inputs into clusters by comparing the inputs amongst themselves

Question 4

What are examples of unsupervised learning problems

Answer 4

• Clustering
• Customer segmentation
• Outlier Checker

Question 5

What is Reinforcement Learning

Answer 5

The goal of reinforcement learning is to train an agent to perform a certain task by learning from interactions with its environment

Question 6

What is Generative Modelling

Answer 6

The goal is to produce data, such as an image or text, given a label or prompt, rather than recognising a label from the data