14 Backpropagation
As we’ve seen, a neural network is just a collection of neurons, each doing its own little calculation and then passing on its results to other neurons. How can we train such a thing to produce the results we want? And how can we do it efficiently?
The answer is backpropagation, or simply backprop. Without backprop, we wouldn’t have today’s widespread use of deep learning because we wouldn’t be able to train big networks in reasonable amounts of time. Every modern deep learning library provides a stable and efficient implementation of backprop. Even though most people will never implement backprop, it’s important to understand ...
Get Deep Learning 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.
