The Machine Learning Race Is Really a Data Race

Re:The Machine Learning Race Is Really a Data Race by zozbot123 on Sunday December 23, 2018 @06:08PM (#18749015)

The general answer is yes - it's the "one shot learning" problem. But humans also have access to a large (if not very well defined) amount of data via background knowledge, which plausibly enables them to apply stronger, more effective priors to any one task.

[ Parent | on HN ]

Re:The Machine Learning Race Is Really a Data Race by lordnacho on Sunday December 23, 2018 @06:18PM (#18749067)

This is the thing. Machines start on a blank slate each time, humans never.

So how do you compare the amounts of data needed?

[ Parent | on HN ]

Re:The Machine Learning Race Is Really a Data Race by ars on Monday December 24, 2018 @03:04AM (#18751057)

So try it: Teach a computer about animals in general.

Then show it a single example of a new animal it has never seen, and see how well it would do at telling you if challenge images are or are not that animal.

It's not even going to be close to what a human would manage.

It's not because of the data, it's because a human understands what he's seeing.

[ Parent | on HN ]

Re:The Machine Learning Race Is Really a Data Race by lordnacho on Monday December 24, 2018 @04:37AM (#18751291)

Then you're comparing a machine that's never seen anything but animals against a human that has no separation between different kinds of data and can't start on a blank slate.

I may have all sorts of images in my mind, allowing me to make the distinction?

[ Parent | on HN ]

Re:The Machine Learning Race Is Really a Data Race by dmurray on Monday December 24, 2018 @01:24PM (#18753446)

This doesn't sound too different from face recognition systems where you train a network to be able to tell if two faces are the same or different. I don't know what the state of the art in animal recognition is, though.

[ Parent | on HN ]