Tarski–Kuratowski algorithm
Appearance
In computability theory and mathematical logic the Tarski–Kuratowski algorithm is a non-deterministic algorithm that produces an upper bound for the complexity of a given formula in the arithmetical hierarchy and analytical hierarchy.
The algorithm is named after Alfred Tarski and Kazimierz Kuratowski.
Algorithm
[edit]The Tarski–Kuratowski algorithm for the arithmetical hierarchy consists of the following steps:
- Convert the formula to prenex normal form. (This is the non-deterministic part of the algorithm, as there may be more than one valid prenex normal form for the given formula.)
- If the formula is quantifier-free, it is in and .
- Otherwise, count the number of alternations of quantifiers; call this k.
- If the first quantifier is ∃, the formula is in .
- If the first quantifier is ∀, the formula is in .
References
[edit]- Rogers, Hartley The Theory of Recursive Functions and Effective Computability, MIT Press. ISBN 0-262-68052-1; ISBN 0-07-053522-1