What is Machine Learning?

Two definitions of Machine Learning are offered. Arthur Samuel described it as: "the field of study that gives computers the ability to learn without being explicitly programmed." This is an older, informal definition.

Tom Mitchell provides a more modern definition: "A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E."

Example: playing checkers.

E = the experience of playing many games of checkers

T = the task of playing checkers.

P = the probability that the program will win the next game.

Machine Learning is used everywhere, you might be benefiting from it without even knowing. For example:

• Facebook photo face recognition on your friends
• Email spam detection
• Web search engines such as Google and Bing

Machine Learning is a field that grew out of A.I. We are capable of writing simple A.I. programs to do tasks such as finding the shortest path between two points. However, we have no idea how to write efficient and accurate programs to perform tasks like facial recognition, email spam detection and web search. Therefore we train machines to learn on their own how to perform those tasks, hence the emergence of Machine Learning popularity.

There are two main applications for machine learning right now. First being database mining. Large online service providers such as ebay, Facebook, Netflix continue to collect large amount of user data. They use machine learning to understand their users better and to serve them better. How else do you think Netflix is able to recommend the next movie for millions of users? Medical research facilities also use machine learning to find correlation in medical records so they can understand disease better and possibly mitigate the onset of some diseases. Machine learning is also applied in genome projects where researchers decode the human gene sequence.

The other application is to use machine learning to perform tasks that we cannot program by hand. Autonomous driving uses Machine Learning to teach computers how to drive. Machine Learning is used in Natural Language Processing so you can speak to a bot as if they are humans. Computer vision uses Machine Learning so they can identify a cat in a picture.

In general, any machine learning problem can be assigned to one of two broad classifications:

Supervised learning and Unsupervised learning.