Overview / Basic Ideas / Limitations

Many traditional problems of computer science require just
one problem-defining input at the beginning of the problem solving
process. For example, the initial input may be a large integer,
and the goal may be to factorize it.
In what follows, however, we will
also consider the *more general case*
where the problem solution requires interaction with a dynamic,
initially unknown environment that produces a continual stream of
inputs and feedback
signals, such as in autonomous robot control tasks,
where the goal may be to maximize expected cumulative
future reward [20].
This may require the solution of essentially arbitrary problems
(examples in Section 6.2 formulate traditional
problems as special cases).

- Set-up and Formal Goal
- Basic Idea of Gödel Machine
- Proof Techniques and an -Optimal Initial Proof Searcher
- Limitations of Gödel Machines

Juergen Schmidhuber 2005-01-03