Optimizing Neural Networks For Deep Learning | Towards Data Science

How to improve training beyond the “vanilla” gradient descent algorithm