OKR

Page 1

Arthur Samuel (1958) - Field of study that gives computers ability to learn without being explicitly programmed.

Tom Mitchell (1998) - A computer program said to learn from Experience E on some task T whose performance can be measured by P, if its performance on T measured by P can be improved by E

Types of Machine Learning

Supervised Learning - User supervise(teach) machine by using set of programs

In supervised learning, we are given a data set and already know what our correct output should look like, having the idea that there is a relationship between the input and the output.

Supervised learning is categorized in basically Regression and Classification problems.

Regression​​​​​​

we predict results from continuous output, that means we are trying to map input variable with continuous functions

Classification

We try to predict results in discrete outputs.

Problem which have different attributes for decision making

Unsupervised Learning - Machine going to learn by itself.

We are given a set of data and we approach problem with little or no idea of what our result will look like.

we can derive this structure by clustering the results based on relationship among variables within data.

Notations

m = No of training examples, x = input variables/features, y = output variable/target

Page 2