Title here is wrong. Title in article and headings in article are right: ONLINE gradient descent
It's specifically not stochastic. From the article:
Online gradient descent
Finally, we have enough experience to implement online gradient descent. To keep things simple, we will use a very vanilla version:
- Constant learning rate, as opposed to a schedule.
- Single epoch, we only do one pass on the data.
- Not stochastic: the rows are not shuffled. ⇠ ⇠ ⇠ ⇠
- Squared loss, which is the standard loss for regression.
- No gradient clipping.
- No weight regularisation.
- No intercept term.