Supervised Learning:

How it works:
Supervised Learning is a foundational concept in Machine Learning where an algorithm learns from Labelled Data.
This means that each input is paired with a known output and remembered by the model.
The goal is to train the model so that it can accurately predict outputs, from new unseen inputs.

Pattern Recognition:
Supervised Learning models are excellent at finding hidden patterns and structures.
It’s uncanny ability to find hidden patterns (Pattern Recognition) makes it an ideal candidate for Medical Imaging Diagnostics.
However, this is heavily reliant upon the training data being clean, well labelled, and vast.

Image Classification:
Using Pattern Recognition, models are able to organise images by similarity, categorising them into their appropriate classes.

Object Detection:
Using Labelled Data models are able to identify objects from their input data.
The detected objects are given a confidence scores, (a percentage of how confident the model is, that the input data matches a particular label).
If the confidence score is greater than a pre set threshold, then the object is surrounded by a bounding box with the appropriate label displayed.

Face Detection:
Face Detection falls under the umbrella of Object Detection but in this case the “object” is a human face.

Face Recognition:
A clear example of Supervised Learning.
The model is trained on a persons face by measuring and recording distances between specific positions.
If it sees those set of measurements again, the face is recognised.

Supervised Learning can be broken down into three phases:
The Training Phase: The model sees many input-output pairs and adjusts its internal parameters to minimize prediction errors.
The Validation Phase: The model is tested on new labelled data to check how well it generalises.
The Prediction Phase: Once trained, the model can predict outputs for new, unlabelled inputs.

Supervised Learning uses two Algorithms:
Regression: Modelling the relationship between input features and an output variable Classifications.
Classification: Assigning input data to predefined categories by learning patterns from labelled examples.