I have created a CNN for use on the MNIST dataset for now (so I have 10 classes). I have trained SVMs on the sublayers of this trained CNN and wish to combine them into a combined SVM as to give a combined score.

So far, I trained two individual SVMs at two of the sublayers of my neural network. What is the best method I can go about combining the two SVMs and what are the different options available to me? Is it simply a case of taking the maximum/average of each SVM prediction for a class and using that as the score for the combined SVM class prediction?


  • $\begingroup$ You can google "deep svm", there are several approaches in papers $\endgroup$ Dec 30 '19 at 6:37

Your Answer

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

Browse other questions tagged or ask your own question.