What are advanced and reverse passages in neural networks?

Question

What is the point of moving back and forth in neural networks?

Everyone mentions these expressions when talking about backpropagation and epochs.

I realized that the transitions back and forth take place together in the era.

Answer 1

I highly recommend you read this article carefully:

http://cs231n.imtqy.com/optimization-2/

It is shown in a fantastic and light unsightly form.

Answer 2

A direct pass refers to the process of calculating the values ​​of the output layers from the input data. It passes through all neurons from the first to the last layer.

.

( ) ( ). , .

.