Re:Large Scale Visual Recognition Challenge 2011 - Results
by jules on Thursday October 04, 2012 @01:20PM
(#4613319)
Yeah, if you use the same raw RGB features for the SVM as the neural net then the neural net would blow the SVMs away even more utterly.
Re:Large Scale Visual Recognition Challenge 2011 - Results
by pmelendez on Thursday October 04, 2012 @01:57PM
(#4613479)
No... but I'd bet that if you use the high dimensional features resulted from the deep convolutional learning process as an input of an SVM the difference would not be that significant.
Re:Large Scale Visual Recognition Challenge 2011 - Results
by jules on Thursday October 04, 2012 @05:56PM
(#4614611)
Well yeah, but then you're basically putting the meat of the NN algorithm into the SVM. I'd call the resulting algorithm a neural network with an SVM frosting. You might as well train naive bayes directly on the final nth layer of the NN instead of SVM on the (n-1)th layer, would be an almost equally weak argument for the thesis that NNs are not superior to the other algorithms on this task, since basically all the power is coming from the NN.
I will be more agree with a title like "Deep Convolutional learning overperformed traditional techniques in Object Recognition"