>> Right, so unsupervised learning we don't get those examples.

We have just essentially something like input, and

we have to derive some structure from them just by looking at

the relationship between the inputs themselves.

>> Right so give me an example of that.

>> So.

When you're studying different kinds of animals, say, even as a kid.

>> Mm-hm. >> You might start to say, oh,

there's these animals that all look kind of the same.

They're all four-legged.

I'm going to call all of them dogs.

Even if they happen to be horses, or cows, or whatever.

But I have developed, without anyone telling me,

this sort of notion that all these belong in the same class.

And it's different from things like trees.

Which don't have four legs.

>> Well some do, but I mean, they have, they both bark, is all I'm saying.

>> Did I really set you up for that?

>> Not on purpose.

>> I, I'm sorry, I want to apologize to each and every one of you for that.

But that was pretty good.

Michael is very good at word play.

Which I guess is often unsupervised as well.

No, I get a lot of [LAUGH].

>> [LAUGH] You certainly get a lot of feedback.

>> Yeah, that's right.

So I say, please stop doing that.

>> So if supervised learning is about function approximation,

then unsupervised learning is about description.

It's about taking a set of data and

figuring out how you might divide it up in one way or the other.

>> Or maybe even summarization it's not just the description but

it's a shorter description.

>> Yeah, it's usually concise.

Compression.

>> Compact description.

So I might take a bunch of pixels like I have here it might say, male.

>> [LAUGH] Wait, wait, wait, wait.

I'm pixels now?

>> As far as we can tell.

>> That's fine.

>> I however, am not pixels.

I know I'm not pixels.

I'm pretty sure the rest of you are pixels.

>> That's right. >> So I have a bunch of pixels, and

I might say male.

And or I might say female.

Or I might say dog.

Or I might say tree.

But, the point is I don't have a bunch of labels that say dog, tree, male,

or female.

I just decide that pixels like this belong with pixels like this.

As opposed to pixels like something else that I'm pointing to behind me.

>> Yeah we're living in a world right now that is devoid of any other objects.

Oh, chairs!

>> Chairs! Right.

So these pixels are very different than those pixels because of

where they are relative to the other pixels.

Say, right?

So, if you were looking-

>> I'm not sure that's helping me understand unsupervised learning.

>> Go out and, go outside and

look at a crowd of people and try to decide how you might divide them up.

Maybe you'll divide them up by ethnicity.

Maybe you'll divide them up by whether they have purposefully shaven

their hair in order to mock the bald or whether they have curly hair.

Maybe you'll divide them up by whether they have

goatees, >> Facial hair.

>> Or whether they have grey hair,

there's lots of things that you might do in order,.

>> Did you just point at me and say grey hair?

>> I was pointing and your head happened to be there.

>> Oh come on. >> Pixels,

>> Where's the grey hair?

>> It's a two dimensional, right there, it's right where your spit curl is.

All right.

>> Okay. So, imagine you're dividing the world

up that way.

You could divide it up male, female.

You could divide it up short, tall, wears hats, doesn't wear hats,

all kinds of ways you can divide it up.

And no one's telling you the right way to divide it up, at least not directly.

That's unsupervised learning.

That's description, because now-

>> Mm. >> Rather than having to send pixels of

everyone, or having to do a complete description of this crowd,

you can say there were 57 males and 23 females say.

Or there are mostly people with beards.

>> So on. >> Or whatever.

>> I like summarization for that.

That seems good.

>> I like summarization for that.

>> Yeah. >> It's a nice concise description.

That's unsupervised learning.

>> Good. Very good.

>> And practice. >> And that's different from

supervised learning?

>> It's different in supervised learning and it's different in a couple of ways.

One way that it's different is all of those ways that we

could have just divided up the world.

In some sense are all equal either.

So I could divide up by sex or

I could divide up by height or I could divide up by clothing or whatever.

And they're all equally good absent some other signal later telling you

how you should be dividing up the world.

But supervised learning directly tells you there's a signal.

This is what it ought to be, and that's how you train.

And those are very different.

>> Now, but I could see ways that unsupervised learning could be helpful

in the supervised setting, right?

So if I do get a nice description, and it's the right kind of description,

it might help me map to, it may help me do the function approximation better.

>> Right, so instead of taking pixels that input, as input, and,

and labels like, male or female.

I could just simply take a summarization of you like how much hair do

you have your relative height, the weight, and

various things like that that might help me do it.

That's right. And by the way,

in practice this turns out to be things like density estimation.

We do end up turning it into statistics at the end of the day.

Often.

>> But it's statistics from the beginning.

But when you say density estimation.

>> Yes. >> Are you saying I'm stupid?

No. >> All right so

what is density estimation?

>> Well they'll have to take the class to find out.

>> I see.

>> Okay