1
$\begingroup$

Is the following statement about neural networks overclaimed?

Neural networks are iterative methods that minimize a loss function defined on the output layer of neurons.

I wrote this statement in the Introduction section of a conference paper. Reviewer got back to me saying that, "this statement is over claimed without proper citations". How is this statement over claimed? Is it not obvious that neural networks are iterative methods that try to minimize a cost function?

$\endgroup$
  • $\begingroup$ Hi and welcome to AI SE! Can you ask a question that can be asked more objectively? It's difficult to objectively answer the current question. Furthermore, note that I've changed the title of this post to be equal to the question in the body. Please, clarify what is your real question. $\endgroup$ – nbro Feb 3 at 22:19
2
$\begingroup$

I think I would never say that neural networks are iterative methods. I would say that iterative methods (e.g. gradient descent) are used to train neural networks (which can be thought of as linear and non-linear models, but mainly non-linear), which is quite different. Maybe you should or wanted to say that deep learning is an area of study where iterative methods are used to train neural networks.

It is possible the reviewer is just telling you that this or similar statements have been used a lot and don't really provide any insight, they are misleading, or simply useless or trivial in the context of the paper or conference.

Without further clarifications by the reviewer and about your paper and conference, it is difficult to provide more explanations.

| improve this answer | |
$\endgroup$
  • $\begingroup$ I think I agree with, "I think I would never say that neural networks are iterative methods". There are many kinds of NN and some(say spiking) use unsupervised methods to train. However, I agree that, "I would say that iterative methods (e.g. gradient descent) are used to train neural networks". $\endgroup$ – Ruthvik Vaila Feb 4 at 1:53
  • $\begingroup$ What does the part "loss function defined on the output layer of neurons" mean? $\endgroup$ – DuttaA Feb 4 at 3:03

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.