Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Large Scale Visual Recognition Challenge 2011 - Results

vision.stanford.edu

Post reply on HN
Print view View original

31–38 of 38 posts

Re: Large Scale Visual Recognition Challenge 2011 - Results

#31

Post by pmelendez »

jules wrote: post #29
pmelendez wrote: post #14

Earlier quoted context omitted.

Sorry for disagree, but it seems more related to the fact that they are using deep convolutional learning rather than the neural network itself. If you use an ANN with the same set of features side by side with a SVM you will see very equivalent results. I will be more agree with a title like "Deep Convolutional learning overperformed traditional techniques in Object Recognition"

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.

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

#32

Post by modeless »

iandanforth wrote: post #24

Hinton's team (SuperVision) uses an interesting 'dropout' technique. He gave a Google Tech Talk on this back in June. http://www.youtube.com/watch?v=DleXA5ADG78&feature=plcp And an older talk that covers some of what a deep convolutional net is: http://www.youtube.com/watch?v=VdIURAu1-aU

Hinton is currently teaching a Coursera class on neural nets: https://class.coursera.org/neuralnets-2012-001/class/index

So far I've watched the first lecture and it seems like it'll be exactly the course I've been wanting: starting with the basics of machine learning but quickly diving into the state of the art for neural nets.

Re: Large Scale Visual Recognition Challenge 2011 - Results

#33

Post by Dn_Ab »

pmelendez wrote: post #11
pjin wrote: post #5

Earlier quoted context omitted.

To nitpick at the math: "No free lunch" results are asymptotic in the sense that they necessarily hold over the _entire_ domain of whatever problem you're trying to solve. Obviously, algorithms will and do perform differently over the relatively few inputs (compared to infinity...) that they actually encounter. It's similar to undecidability: just because a problem is generally undecidable doesn't mean you can't comp…

Agreed... I was in a rush to catch the train this morning and I didn't have chance to elaborate, I shouldn't do that. However, my point was that most of the algorithms used on that link (ANN, SVM, etc) had similar expressive power (VC dimension) and had been proved to have similar performance between them in object recognition. People normally take advantage on their specific properties rather than paying too much at…

Although SVMs and layered neural nets have similar expressivity, the similarity is very much like turing completeness. i.e. Can't tell aparts the haskells from the unlambdas. SVMs express certain functions in a manner that grows exponentially with input vs a deep learner which tends to be more compact. The key to being a deep learner is in using unsupervised learning to seed a hierarchy of learners learning ever more abstract representations.

Also, Multilayered Kernel learners already exist.

Re: Large Scale Visual Recognition Challenge 2011 - Results

#34

Post by modeless »

jules wrote: post #27
sumodds wrote: post #6

Am not sure if you can apply winner takes all for such marginal difference in error. Give a slightly different database and things go awry. Check out : "Unbiased Look at Dataset Bias", A. Torralba, A. Efros,CVPR 2011.

The difference in error between the first and the rest is ENORMOUS. Task 1: 1st 0.15315 (convolutional neural net) 2nd 0.26172 3rd 0.26979 4th 0.27058 5th 0.29576 [...] Differences: 0.10857 0.00807 0.00079 0.02518 As you can see the first is way ahead of the rest. The difference between the 1st and 2nd is ~11%, between the second and third ~1%. Task 2: 1st 0.335463 (convolutional neural net) 2nd 0.500342 3rd 0.536474…

the results were obtained with a relatively general purpose learning algorithm. No extraction of SIFT features, no "hough circle transform to find eyes and noses".

This deserves even more emphasis. All of the other teams were writing tons of domain specific code to implement fancy feature detectors that are the results of years of in-depth research and the subject of many PhDs. The machine learning only comes into play after the manually-coded feature detectors have preprocessed the data.

Meanwhile, the SuperVision team fed raw RGB pixel data directly into their machine learning system and got a much better result.

Re: Large Scale Visual Recognition Challenge 2011 - Results

#35

Post by pmelendez »

Dn_Ab wrote: post #33
pmelendez wrote: post #11

Earlier quoted context omitted.

Agreed... I was in a rush to catch the train this morning and I didn't have chance to elaborate, I shouldn't do that. However, my point was that most of the algorithms used on that link (ANN, SVM, etc) had similar expressive power (VC dimension) and had been proved to have similar performance between them in object recognition. People normally take advantage on their specific properties rather than paying too much at…

Although SVMs and layered neural nets have similar expressivity, the similarity is very much like turing completeness. i.e. Can't tell aparts the haskells from the unlambdas. SVMs express certain functions in a manner that grows exponentially with input vs a deep learner which tends to be more compact. The key to being a deep learner is in using unsupervised learning to seed a hierarchy of learners learning ever more…

"The key to being a deep learner is in using unsupervised learning to seed ..."

Exactly! that was my whole point which doesn't makes sense now that the title had changed.

"Also, Multilayered Kernel learners already exist"

I didn't know that and I'll check that shortly, thanks for the info.

Re: Large Scale Visual Recognition Challenge 2011 - Results

#36

Post by sumodds »

jules wrote: post #27
sumodds wrote: post #6

Am not sure if you can apply winner takes all for such marginal difference in error. Give a slightly different database and things go awry. Check out : "Unbiased Look at Dataset Bias", A. Torralba, A. Efros,CVPR 2011.

The difference in error between the first and the rest is ENORMOUS. Task 1: 1st 0.15315 (convolutional neural net) 2nd 0.26172 3rd 0.26979 4th 0.27058 5th 0.29576 [...] Differences: 0.10857 0.00807 0.00079 0.02518 As you can see the first is way ahead of the rest. The difference between the 1st and 2nd is ~11%, between the second and third ~1%. Task 2: 1st 0.335463 (convolutional neural net) 2nd 0.500342 3rd 0.536474…

Lol.. my bad. I did not pay attention. I thought the error was in percentages. (I was comparing with MNIST and somehow assumed this too was percentages). Come to think of it, that is really dumb (what that would mean) !!

Re: Large Scale Visual Recognition Challenge 2011 - Results

#37

Post by jules »

pmelendez wrote: post #31
jules wrote: post #29

Earlier quoted context omitted.

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.

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.

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.

Re: Large Scale Visual Recognition Challenge 2011 - Results

#38

Post by sumodds »

Evbn wrote: post #19
sumodds wrote: post #6

Am not sure if you can apply winner takes all for such marginal difference in error. Give a slightly different database and things go awry. Check out : "Unbiased Look at Dataset Bias", A. Torralba, A. Efros,CVPR 2011.

I worry you may have taken a biased look at "Unbiased Look at Dataset Bias".

Not that only, I had a high variance on my bias.. ;)
Post reply on HN