Sorry, I'm really new to neural networks and this question is probably pretty obvious. If you have any resource that can help clarify these concepts to me it would be much appreciated.

The way I understand neural networks is that each input node, x i goes to each hidden layer. Is there a unique weight for each input that goes into each hidden node?

For example, say I have this neural net

with xi being inputs, h being hidden layers, and O being the output. Would the weight from x1 to h1 be different from the weight from x1 to h2? If I'm not on the right track, please correct me!

  • $\begingroup$ You're on the right track, but this question is much too basic to warrant a full answer. My suggestion, in this case, would be to first ask in Computer Science Chat if anyone can direct you to a suitable reference. $\endgroup$ Aug 19, 2014 at 2:06
  • 1
    $\begingroup$ New users can't use chat. @elder4222 Coursera.org has a couple of online machine learning courses that cover neural nets, linear classifiers and related topics. You might find one of these courses more engaging than just buying a machine learning textbook and slogging through it. $\endgroup$
    – Kyle Jones
    Aug 19, 2014 at 3:44
  • $\begingroup$ @KyleJones thanks, I'll be sure to look at it. $\endgroup$
    – elder4222
    Aug 19, 2014 at 3:45

1 Answer 1


Yes, each input value is weighted before being fed into the nodes in the next layer. Each input/output pair of "neurons" has its own weight and it is these weights that are adjusted by the neural network training process. The initial weights are typically random values before training begins.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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