Basic Models for Supervised Learning

(28th-May-2020)

A learned model is a representation of a function from the input features to the target features. Most supervised learning methods take the input features, the target features, and the training data and return a model that can be used for future prediction. Many of the learning methods differ in what representations are considered for representing the function. We first consider some basic models from which other composite models are built. Section 7.4 considers more sophisticated models that are built from these basic models.

1. Learning Decision Trees

2. Searching for a Good Decision Tree

3. Linear Regression and Classification

4. Squashed Linear Functions

5. Bayesian Classifiers