0
$\begingroup$

Is it possible to consider an RNN as a classical feedforward neural network that just take the precedent output as a part of the input ?

$\endgroup$
0
$\begingroup$

Almost. I think that to match the common interpretation of an RNN you need to also have a new input at each time-step (whereas you used the word "just" suggesting otherwise).

What you're describing is some function $f_1: \mathbb{R}^n \to \mathbb{R}^n$ whereas the "conventional" RNN is more like $f_2: \mathbb{R}^m \times \mathbb{R}^n \to \mathbb{R}^n$, where $\mathbb{R}^m$ refers to the new information at each time step.

Of course, this is all an exercise in interpreting semantics. So the other answer to your question is: "if you want it to"

$\endgroup$

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.