In computability theory, a set of natural numbers is called computable, recursive, or decidable if there is an algorithm which takes a number as input, terminates after a finite amount of time (possibly depending on the given number) and correctly decides whether the number belongs to the set or not.

A set which is not computable is called noncomputable or undecidable.

A more general class of sets than the computable ones consists of the computably enumerable (c.e.) sets, also called semidecidable sets. For these sets, it is only required that there is an algorithm that correctly decides when a number is in the set; the algorithm may give no answer (but not the wrong answer) for numbers not in the set.

Formal definition

A subset of the natural numbers is called computable if there exists a total computable function such that if and if . In other words, the set is computable if and only if the indicator function is computable.

Examples and non-examples

Examples:

Non-examples:

Main article: List of undecidable problems

Properties

If A is a computable set then the complement of A is a computable set. If A and B are computable sets then AB, AB and the image of A × B under the Cantor pairing function are computable sets.

A is a computable set if and only if A and the complement of A are both c.e. The preimage of a computable set under a total computable function is a computable set. The image of a computable set under a total computable bijection is computable. (In general, the image of a computable set under a computable function is c.e., but possibly not computable).

A is a computable set if and only if it is at level of the arithmetical hierarchy.

A is a computable set if and only if it is either the range of a nondecreasing total computable function, or the empty set. The image of a computable set under a nondecreasing total computable function is computable.

References