Member-only story
Overview of Human Learning & Machine Learning
A general definition of Machine Learning is:
“Machine learning is the field of study that gives computers the ability to learn without being explicitly programmed.”
— Arthur Samuel, 1959
So, Machine Learning is the science (and art) of programming computers so they can learn from data.
To understand this concept, let us look at how a programmatic solution(Classic Algorithm) differs from a machine learning solution.
In the Classic approach, we write the program on a computer and provide input to it. Then the program runs and produces an output.
In the ML approach, we feed the data and an example of output (Input-Output pairs in the form of a Dataset) into the computer. Here, the programmer does not need to write the program(or very little program); rather, the computer itself learns from the given input-output pairs and returns the program/model.
Example: Spam Filter
If you want to build an ML algorithm to detect whether an email is spam or ham (not spam), you have to provide some existing examples of spam and ham emails. Input: EmailsOutput: Whether the mail is spam or ham
By learning from these input-outputs, a model will be developed that will…