Okay.

Quizzes Alba.

I want to talk about a couple of numbers I've talked before, but I'm still worried.

Okay.

We have something like 600 people on Stodon.

I know this means nothing.

But a number that means something is 280 of you have registered for the exam.

Somewhere between 60 and 100, probably depends on the weather, are doing the quizzes.

And 10 to 15 are here on average.

Those two numbers are positively correlated with passing the exam.

And I apologize to you because you're not the one I should be preaching to.

So I'm preaching to the Zoomies and the people who are not watching.

That's especially difficult to preach to.

I'm worried about this number.

I'm worried if you're not prepared and if you're not doing the quizzes,

chances are very good that you're not prepared.

So think about what you're doing.

There's a long summer, of course.

The course notes are very complete.

There's the videos.

All the quizzes will be published and can be used and kind of studied and be used for preparation.

So can all the old exams and so on.

But just by watching the videos and nodding, you're not going to pass.

And I would like to have everybody pass.

Right?

I like to give good grades.

But sometimes I can't by what you write in the exam.

So please think about these numbers and maybe think about whether moving from that number to

this number or even that number might actually do you good.

The numbers suggest it.

Not for everybody.

There are people who best study alone by reading.

But on average, attending class, even if you can't play me double speed,

actually works for most people.

Okay.

End of rant.

But I just get worried.

Okay.

We've talked about artificial neural networks.

A very important part of modern AI has been hugely successful.

So successful that everybody wants to give a course about it.

So I'm only going to cover the extreme basics in this course.

Remember, we are the AI one and two classes are kind of look at everything,

wander through the vegetable garden, sniff at every plant, and see what's out there.

So basics of neural networks.

Generally, neural networks is now classified as its own kind of computing paradigm.

Whereas normal computers, things like that for Norman architectures, they basically have

have one complex computing engine or maybe if you have a newer laptop, 16 cores or something like this.

Which operate at huge clock speeds.

Billions of operations per second.