Learning Paradigms in Machine Learning

As we have learnt about Artificial Intelligence, Machine learning and deep learning in the previous article I thought we can look into some learning paradigms. There are 3 main types of learning in machine learning:

Supervised learning:

- In supervised learning, a machine learning model infers a function from labelled training data.

-It aims to learn a function that, given a sample of data and desired outputs, approximates a function that maps inputs to outputs.

-It is usually used in classification and regression

- Common algorithms in supervised learning include logistic regression, naive bayes, support vector machines, artificial neural networks, and random forests.

- In both regression and classification, the goal is to find specific relationships or structure in the input data that allow us to effectively produce correct output data.

Semi-supervised Learning:

-It falls in between Supervised learning and Unsupervised learning.

-The most basic disadvantage of any Supervised Learning algorithm is that the dataset has to be hand-labelled either by a Machine Learning Engineer or a Data Scientist. This is a very costly process, especially when dealing with large volumes of data.

-Semi-supervised learning is used to label unlabeled data points using knowledge learned from a small number of labelled data points.

-This is often used in real-world as labelling data is very expensive and/or usually, it is a constant flow of data.

Unsupervised Learning:

- Unsupervised learning is the training of machine using information that is neither classified nor labelled and allowing the algorithm to act on that information without guidance.

-It learns patterns from unlabeled data. Often look for a structure.

- Unlike supervised learning, no teacher is provided that means no training will be given to the machine. Therefore, the machine is restricted to find the hidden structure in unlabeled data by our-self.

-There are two types of unsupervised learning:

i. Clustering: A clustering problem is where you want to discover the inherent groupings in the data, such as grouping customers by purchasing behaviour.

ii. Association: An association rule learning problem is where you want to discover rules that describe large portions of your data, such as people that buy X also tend to buy Y.

-The disadvantage of any Unsupervised Learning is that its application spectrum is limited.