In 1998/99 I did a Research M.Sc. in Computer Vision Engineering at King's College London. The research topic was Classification of Radar Targets based on their Dynamic Behaviour.
For my B.Sc. I studied Artificial Intelligence and Mathematics at the University of Edinburgh, Scotland.
In my 4th year project I investigated the Dynamic Behaviour of Self-organising Neural Networks built on Marr's Theory of the Neocortex. Interested ? Then have a look at the proposal, the dissertation or an simulation example.
Over the summer 97 I had the opportunity to experience the life of a
full-time mathematician since I got a Faculty Vacation Scholarship.
There were many interesting things to chose from. Investigating
random walks on Fractals or how plants spread and what can
be said about their initial conditions from their final distribution
and more.
I decided to look at Optimal Stopping Time Problems
which occur in general when someone has to make a decision based
only on the information gathered so far and this decision can turn
out to be the wrong one later. If the decision is done too late,
good opportunities might have been missed, while if the decision is
done too early the final outcome might have been judge wrongly.
Therefore one would like to know when the best time for stopping
collecting information and for decision-making has come.
To illustrate the problematic I wrote a report which gives examples of
optimal stopping time problems and explains how they can be solved
mathematically. The chosen examples have in common that the powerful
idea of backward induction plays a key issue how the solutions were
obtained.
Click here for my report about Optimal Stopping Time Problems.
Before all that I worked for Siemens AG in Munich for 12 years in software development.
During this I came across many programming languages and
operational systems.
At the beginning we worked with punch cards,
Assembler and an operating system (HMP1116) which was loaded
from magnetic tape each time.
Nevertheless we managed to build an reliable Low Level Air Control
Radar System.
Then came the time of Pascal and Fortran on Siemens R30 machines. The
source code was now on discs - what leap forward.
It followed 8086 Assembler building test equipment software on
Siemens Microprocessor Equipment (SME) system.
For creating and maintaining databases on BS2000 (operational system
from Siemens) and DOS, I learned the appropriate SQL-languages.
The last years before my studies I was involved in the EuroFighter Project,
doing test and integration for the Radar Data Generator which simulates
the environment for the EuroRadar. This time the operating systems
VME from DEC and EPLX from Control Data were used and the programming
language was ADA.
Finally the programming languages I learned at University are
C, PROLOG, ML and the common operating system is Unix.
For mathematical investigations I use Maple and MATLAB.