Introduction to Machine Learning

Machine learning algorithms can be broadly divided into three categories depending on the nature of the “learning” process, namely, supervised, unsupervised, and reinforcement learningReinforcement learning. In this chapter, we introduce these different categories with the focus on the nature of the tasks for which these algorithms are useful. Specifically, we focus on supervised and unsupervised learningUnsupervised learning algorithms. The supervised algorithms may further be classified as parametric and non-parametric algorithms depending on the mathematical modelModels used to fit the data. These algorithms may be used for several downstream tasks such as classification, regression, or clustering. Finally, we discuss the idea of overfitting and underfitting in machine learningMachine learning algorithms.