Recursive isomorphism

from Wikipedia, the free encyclopedia

In computability theory, recursive isomorphism is an equivalence relation on sets of natural numbers .

definition

Sets of natural numbers are called recursively isomorphic if there is a total computable bijection such that .

Numbering of a (at most) countable set is called recursively isomorphic if there is such a bijection with .

The mapping is then called recursive isomorphism .

properties

  • Recursive isomorphism is an equivalence relation on the power set of the natural numbers.
  • If there is a recursive isomorphism, its inverse function must also be computable.
  • A set is recursively isomorphic to the holding problem if and only if it is creative .
    • In addition, according to the sentence below , these are exactly the RE - complete sets.

Myhill's isomorphism theorem

The following theorem by John Myhill provides a characterization of the concept of recursive isomorphism:

If again be sets of natural numbers, then:

So two sets are recursively isomorphic if and only if they are one-one-equivalent .

The proof of this theorem is an effective variant of the proof of the Schröder-Bernstein theorem .

In the theory of Turing degrees , another important equivalence can be deduced with the help of the isomorphism theorem:

Two sets are in the same Turing degree if and only if their respective Turing jumps are recursively isomorphic.

literature

  • J. Myhill: Creative Sets . In: Journal for Mathematical Logic and Fundamentals of Mathematics Vol. 1 (1955), pp. 97-108.
  • H. Rogers: Theory of recursive function and effective computability (2nd ed.). 1987, MIT Press, Cambridge, MA.