Machine learning is a broad field with various approaches. The three main types are supervised learning, unsupervised learning, and reinforcement learning. Each type has unique characteristics and uses, and they are applied using different algorithms. 

Supervised learning is a common approach where a model is trained using input vectors and their corresponding target vectors. This type of learning relies on labelled data, where each input is associated with a known output. The goal is to learn a mapping from inputs to outputs. Supervised learning is often used in classification problems, where the aim is to assign each input to one of several discrete categories. 

For example, it can be used in digit recognition, where the model is trained to recognize handwritten digits. It is also used in regression problems, where the goal is to predict continuous target variables. An example of this is predicting chemical yield based on reactant concentrations, temperature, and pressure. Some basic algorithms associated with supervised learning include support vector machines (SVMs) and regression techniques