Gradient Descent Algorithm in Java

How can we make a machine learn from data?

Then, how can we make the machine predicts things based on that learned data? Those are the question answered by one of the most classic Machine Learning Algorithms, the Gradient Descent Algorithm, from a Mathematical-Statistical side it’s called Univariate Linear Regression.

This is one of the tools of the Machine Learning toolbox, and what it tries to do is to model a relationship between a scalar dependent variable Y and a explanatory variable X.

In Layman’s term…

Let’s suppose you have a few points distributed in a Graph, so you already know that in a point A you have a well defined X and Y, which means, if you input X, your output will be Y, and in a point B you have a well defined X’ and Y’ as well. But, thing is, if a point emerge between A and B, and you only have the X… what will be the Y (the output)?