The following content has been provided by the University of Erlangen-Nürnberg.

Welcome everybody to this academic year's, lots of genitives, course on artificial intelligence.

This semester we're doing AI-1 in English for the first time.

There are a couple of, or apparently there are a couple of exchange students who can only do English in the audience.

And since my slides are on English, I'll speak English as well.

This is a course which is two semesters.

We're doing symbolic AI, also known as good old-fashioned AI, this semester. And we do statistical AI,

otherwise known as machine learning and uncertainty or statistical AI, next semester.

If you look into the media, when somebody says AI, that's what they mean.

Which only means they have no clue.

First things first, you're too many for the room. Happens every semester.

First time it was much, much worse because we only had a room with 50 seats.

We had people standing in the hallway.

The course is being televised, as you see. So you can actually, if you don't find a seat and don't want to stand,

you can actually look at it two days later or something like this.

Somewhere on your laptop, in the comfort of your own room.

Best with friends in public viewing or whatever you prefer.

Typically, after a couple of lectures, the crowd that's actually here diminishes.

So if you're frustrated because you have to stand, come back in two weeks or be a little bit more early,

then you'll find a place to sit.

Today, as you probably imagine, we're going to essentially do admin.

Grading, what's up with the exam, how are we organising things,

where can you find the relevant things like the course notes and so on.

Maybe a little bit about myself.

I have a professorship for knowledge representation and knowledge processing here at the computer science department.

We're doing research in that direction.

I'm going to tell you a little bit about that because one of the things I want to do,

giving this course on AI, is to recruit new friends for our research.

So if you're interested, AI is what we do, but of course, as always, we're not doing all of AI.

We're doing just a little teensy weensy little bit of AI.

Because AI is an extremely diversified and specialised area,

and in the beginning, 50 years ago, 60 years ago, almost 70 years ago, you could do everything.

Because AI was about this big.

Now AI is much bigger, funding in the billions.

You can imagine that you can't do everything anymore.

Don't think that only because I'm telling you about constrained satisfaction or Bayesian networks

or machine learning in this course that I'm doing research or my group even is doing research in this direction.

We're not. I'm going to tell you what we're doing as we go along.

So are there any prerequisites?

And the answer is, kinda.

So since this is a course that's in the fifth semester of the Bachelor's studies,

but traditionally attracts many master's students,

I'm kind of building on the base curriculum of computer science at FIO.

Something like algorithms and data structures is going to be important.

Because we're going to build on data structures and we're going to talk about algorithms.

That's usually not the problem.

Then I'm assuming a passing familiarity with logic.

Something like Gloin.

But since I'm a logician, I'll be very happy to talk about logic in a more foundational way.

So if you have questions, please ask.