Why is the representation of hidden layers made in terms of circles?

What is the difference between a circle and a box in diagrams of neural networks?

  • 4
    $\begingroup$ There is no fully standardised representation of neural network architecture. It will depend on where you have seen this. Some writers might want to show a key difference by changing the shape of the node. Others will not care and will just show circles or boxes for everything. Could you please show an example of where you have seen this? $\endgroup$ – Neil Slater Feb 27 '19 at 8:19

I guess the reason is that we call them nodes and nodes are usually depicted using circles in graph theory. We model them as graphs for both forward and backward paths.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.