0
$\begingroup$

Can anyone help on how can I train the neural networks with back-propagation using MATLAB? I've tried to train my data with its neural network toolbox but I can't find the Back-propagation option for training data.

$\endgroup$
1
  • $\begingroup$ Eric Wan has a "diagrammatic" method that will save you from horrible chain rules. Try starting here. Once you know the gradient then you can use gradient descent methods. If you are looking for "push the button" instead of "understand the guts" then I'm sure MathWorks has tutorials on the subject. Is this self-study related? $\endgroup$ Jun 23 '17 at 19:28
1
$\begingroup$

Backpropagation is a fast way to compute gradients, which are then used in the optimization algorithm. All of MATLAB's training algorithms probably use backpropagation under the hood to compute the gradients. The only difference between the algorithms is how they then use the gradients. You can read more about their algorithms here: https://www.mathworks.com/help/nnet/ug/train-and-apply-multilayer-neural-networks.html

$\endgroup$

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.