Semi-supervised

Question 1

What is Semi-Supervised learning?

Answer 1

A machine learning approach that uses a small amount of label data and a large amount of unlabelled data to train a model

Question 2

Role of labeled data

Answer 2

Labeled data provides model with guidance teaching it what the correct output should looks like.

Question 3

Role of unlabelled data

Answer 3

Unlabelled data helps the model refine and generalise what it learned from the small labeled dataset, making it more accurate and scalable

Question 4

Why use semi-supervised learning?

Answer 4

Labelling data is expensive and time consuming

Most real world data is un label

Semi supervised learning combines the accuracy of supervised learning with the scalability of unsupervised learning

Question 5

Example of Semi-Supervised learning

Answer 5

Image recognition: A photo app may use a few label image (cat or dog) and unlabelled ones to improve it ability to recognises animal in new photo