UTM theorem

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

In computability theory, the UTM theorem, or universal Turing machine theorem, is a basic result about Gödel numberings of the set of computable functions. It affirms the existence of a computable universal function, which is capable of calculating any other computable function.[1] The universal function is an abstract version of the universal Turing machine, thus the name of the theorem.

Roger's equivalence theorem provides a characterization of the Gödel numbering of the computable functions in terms of the smn theorem and the UTM theorem.

Theorem

[edit | edit source]

The theorem states that a partial computable function u of two variables exists such that, for every computable function f of one variable, a Gödel number e exists such that f(x)u(e,x) for all x. This means that, for each x, either f(x) and u(e,x) are both defined and are equal, or are both undefined.[2]

The theorem thus shows that, defining φe(x) as u(e, x), the sequence φ1, φ2, ... is an enumeration of the partial computable functions. The function u in the statement of the theorem is called a universal function.

References

[edit | edit source]
  1. ^ Rogers 1987, p. 22.
  2. ^ Soare 1987, p. 15.
  • Lua error in Module:Citation/CS1/Configuration at line 2172: attempt to index field '?' (a nil value).
  • Lua error in Module:Citation/CS1/Configuration at line 2172: attempt to index field '?' (a nil value).