Building high-level features using large scale unsupervised learning

Re:Building high-level features using large scale unsupervis... by mturmon on Friday June 22, 2012 @11:50AM (#4147274)

I'm not a large scale ML person, and not intending to take away from the achievement of the team in the OP, but experiments in large scale, unsupervised learning have been going for a long time (even using the autoencoder approach). When you think about it, large scale requires unsupervised...

Here is an old example with hundreds of millions of records and instances:

http://aaaipress.org/Papers/KDD/1998/KDD98-028.pdf

Both authors are now with Google.

Also, people here may not be as up to speed on the state of the art in face rec as they think they are. It's not as much of an unsolved problem as it was even 10 years ago.

[ Parent | on HN ]

Re:Building high-level features using large scale unsupervis... by JohnHaugeland on Friday June 22, 2012 @02:26PM (#4148023)

"When you think about it, large scale requires unsupervised..."

Not necessarily. Crowdsourcing is another option, like Google's image tagging game, reCAPTCHA, et cetera.

Pay a herd of people to do things, and they'll do things for you. You don't have to pay them in money. Telling them they have a high score is often enough.

[ Parent | on HN ]

Re:Building high-level features using large scale unsupervis... by mturmon on Friday June 22, 2012 @03:40PM (#4148378)

Yes, "requires" was too strong. I should have said they go well together. I was trying to get at the fact that it's highly common for large-scale work to be unsupervised.

[ Parent | on HN ]

Re:Building high-level features using large scale unsupervis... by marshallp on Friday June 22, 2012 @02:01PM (#4147891)

Face recognition usually uses a hand-coded layer followed by a machine learning algorithm. This technique automatically devises that hand-coded layer. It also did this for 20,000 other categories and can also be applied equally well to audio or any other data type. Huge difference.

[ Parent | on HN ]

Re:Building high-level features using large scale unsupervis... by mturmon on Friday June 22, 2012 @03:43PM (#4148387)

The large-number-of-categories result was the most novel and surprising to me.

[ Parent | on HN ]