So good morning also from my side.

In this winter term now we will speak about the mathematical foundations about artificial

intelligence neural networks and data analytics.

So this is the first part of a whole lecture on neural networks.

But this time period here we will focus on feedforward neural networks only.

So let me make one comment on the title here.

Mathematical foundations means I do not speak to a computer.

I speak to you, which means I want to have your understanding.

It's not only the case that you know an algorithm that can run on a computer.

I want to teach you the understanding why these algorithms are formulated the ways how

they are formulated at the end.

So the inside is the important point for me.

And what is the content of this insight?

Artificial intelligence is a vision.

So this is the idea to make computers similar to what human beings can do.

Neural networks is a type of mathematics which should realize this vision of artificial intelligence.

Now at the end of the day you have to do something practical.

And this practical means you have to do data analytics there.

So it's a sequence of points.

And this sequence of points we have to highlight from the different views.

And as I told you before, in this semester we will speak about feedforward neural networks.

Now if you think about this, then you could say, this is all on a lecture.

And so this is on many hundred slides.

And you can say, if I have the slides, everything is OK.

No, it's not OK.

The real point is that you have an understanding that you have an insight.

So at the end of the term, the point is you as a person have to be able to think in this

style what I'm teaching here.

It's not only that you have slides or that you have the video.

Now let me say something about myself.

The obvious information downside there is I'm affiliated to Fraunhofer Society.

But this was not always the case.

And why is this not going on?

Yeah, here we are.

So at the moment I work for Fraunhofer.

But I started also as a student and I did study mathematics at the University of Bonn.

At this time, there was no study on artificial intelligence.

There was no study on neural network stuff.

So the point is my study was on mathematics with a focus on control theory.

And later I did a PhD in mathematical finance.

This was from the mathematical viewpoint, this was game theory, which is an extension

to control theory.

Then from 1987 to November 2017, I worked in the corporate technology department of

Siemens in Munich.

So this is the central research and development department of Siemens.

And there I was one of the founders of neural network research at Siemens.

So this is because in 1987 Siemens thought about is this artificial intelligence, is

this neural network stuff something which Siemens should do?

And I was one of the persons who have written a proposal to say, yes, Siemens should do

so and in which way we should do so.