Supervised Learning in Machine Learning

Supervised Learning

These algorithm generates a function that maps inputs to desired outputs. One standard formulation of the supervised learning task is the classification problem. The learner is required to learn a function which maps a vector into one of several classes by looking at several input-output examples of the function.

The goal of supervised learning is to learn a mapping from x and y given a training set made of pars (x_i, y_i). Here, the y_i ∈ y are called the labels or targets of the examples x_i. There are two families of supervised learning algorithms:

1. Generative for Supervised Learning: These algorithms try to model the class-conditional density p(x|y) by some unsupervised learning procedure. A predictive density can then be infereed by applying Bayes theorem:

p(x|y)p(x|y)p(y) / ∫_y p(x|y)p(y)dy

2. Discriminative Algorithms: These algorithms don’t try to estimate how the x_i have been generated, but instead concentrate on estimating p(y|x). Some discriminitive methods even limit themselves to modeling whether p(y|x) is greater than or less than 0.5. It is an example of this is the support vector machine (SVM). It has been argued that discriminitive models are more directly aligned with the goal of supervised learning and therefore tend to be more efficient in practice.