Weihrauch reducibility

In computable analysis, Weihrauch reducibility is a notion of reducibility between multi-valued functions on represented spaces that roughly captures the uniform computational strength of computational problems.[1] It was originally introduced by Klaus Weihrauch in an unpublished 1992 technical report.[2]

Definition

edit

A represented space is a pair   of a set   and a surjective partial function  .[1]

Let   be represented spaces and let   be a partial multi-valued function. A realizer for   is a (possibly partial) function   such that, for every  ,  . Intuitively, a realizer   for   behaves "just like  " but it works on names. If   is a realizer for   we write  .

Let   be represented spaces and let   be partial multi-valued functions. We say that   is Weihrauch reducible to  , and write  , if there are computable partial functions   such that where   and   denotes the join in the Baire space. Very often, in the literature we find   written as a binary function, so to avoid the use of the join.[citation needed] In other words,   if there are two computable maps   such that the function   is a realizer for   whenever   is a solution for  . The maps   are often called forward and backward functional respectively.

We say that   is strongly Weihrauch reducible to  , and write  , if the backward functional   does not have access to the original input. In symbols: 

See also

edit

References

edit
  1. ^ a b Brattka, Vasco; Gherardi, Guido; Pauly, Arno (2021), Brattka, Vasco; Hertling, Peter (eds.), "Weihrauch Complexity in Computable Analysis", Handbook of Computability and Complexity in Analysis, Cham: Springer International Publishing, pp. 367–417, arXiv:1707.03202, doi:10.1007/978-3-030-59234-9_11, ISBN 978-3-030-59233-2, S2CID 7903709, retrieved 2022-06-29
  2. ^ Weihrauch, Klaus (1992). The Degrees of Discontinuity of some Translators between Representations of the Real Numbers (Report). Informatik-Berichte. Vol. 129. FernUniversität in Hagen.