Diagram 1 - draw.io (Major Project)

Diagram: Types of Machine Learning

In a machine learning diagram, various types of algorithms are classified based on their characteristics and learning approaches. This categorization helps in identifying the key categories found in the diagram. Here's we can see an overview of the most common categories:

Supervised Learning refers to a type of machine learning algorithm that acquires knowledge from labelled training data. In this method, each data instance has input features and known output labels or targets. The primary objective of such algorithms is to make predictions or classifications on new, unseen data by generalizing the labelled examples. It is also divided into two steps, classification and regression.

In Reinforcement learning, an agent engages with an environment and utilizes trial and error to maximize a reward signal. The agent performs actions within the environment, receives either rewards or penalties as feedback, and modifies its actions accordingly.

Unsupervised learning involves algorithms that analyze unlabeled data, meaning there are input features but no accompanying output labels or targets. The goal of these algorithms is to uncover any underlying patterns, structures, or relationships within the data without any prior knowledge. This type of learning is typically separated into two main categories: clustering and dimensionality reduction.