I'd just like to note that instead of creating additional animosity between SVMs and deep nets, you could use both together. SVMs with hinge loss can be Yet-another-layer (tm) in your deep net, to be used when it provides better performance.
That's a great point. Fundamentally, if you look at something like a CNN, what it's really doing is producing a feature descriptor based on the input image. One can easily use that feature descriptor in a classic SVM, alongside (or instead of) SoftMax.
Yup, in fact, the universal feature extraction is what allows imagenet pretraining to work well on lung cancer images.
One nitpick though, ConvNets can absolutely be used to do "thinking" and more than just feature extraction. For example, fully convolutional networks can be extremely competitive with FC-layer based nets.
Could you explain in a bit more detail how you would integrate an SVM layer into a DNN? The kernel matrix depends on all samples, while at training time you would only have access to those in the minibatch.
The simplest is to pop it on the top. Run you DNN to reduce your input down to a nicer cleaner smaller dimensional output, then plop an SVM on top for classification.
Seems like in that case you would train both models separately on different cost functions. By phrasing it as a layer I was expecting both the SVM and the DNN could be trained simultaneously.
