In the chapter on Stochastic Gradient Descent, we treated the computation of gradients of the loss function as a black box. In this chapter we open this black box and cover the theory and practice of automatic differentiation. Automatic differentiation is a mature technology that allows for the effortless and efficient computation of gradients of arbitrarily complicated loss functions. This is critical when it comes to minimizing loss functions of interest; at the heart ...
© Nikhil Ketkar 2017
Nikhil Ketkar, Deep Learning with Python, https://doi.org/10.1007/978-1-4842-2766-4_9
9. Automatic Differentiation
Nikhil Ketkar1
(1)Bangalore, Karnataka, India
Get Deep Learning with Python: A Hands-on Introduction now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.