• SGD to minimize loss function 
• It assigns the weights in such a way to make the loss as close to zero as possible

• What is the objective of Gradient Descent?

  Gradient, in plain terms means slope or slant of a surface. So gradient descent literally means descending a slope to reach the lowest point on that surface. Let us imagine a two dimensional graph, such as a parabola in the figure below.

   

GD: all of the data points are loaded to find loss/weight 

SGD: one by one points are loaded to find loss/weight

mini batch GD: Points are loaded in batches