In this session, we will go through the algorithm of gradient boosting in a regression setting. With the intuition learned from the previous videos, let us see how this translates to the algorithm behind the model.

Now that you have gone through the steps of Gradient Boosting algorithm in parts, let us summarize these steps in a more structured form.

At any iteration t, we repeat the following steps in the Gradient Boosting scheme of things:

1 Initialize a crude initial functionÂ F0 asÂ $$//argmin\;{\textstyle\sum_{t=i}^T}\;L\left(Y_{I,}\overbrace Y\right)//$$

- This will be the average value which will form our first prediciton.

- For m = 1 to M (where M is the number of trees)

- Calculate the pseudo-residuals.