Cone avoidance and randomness preservation

https://doi.org/10.1016/j.apal.2015.03.001Get rights and content
Under an Elsevier user license
open archive

Abstract

Let X be an infinite sequence of 0's and 1's. Let f be a computable function. Recall that X is strongly f-random if and only if the a priori Kolmogorov complexity of each finite initial segment τ of X is bounded below by f(τ) minus a constant. We study the problem of finding a PA-complete Turing oracle which preserves the strong f-randomness of X while avoiding a Turing cone. In the context of this problem, we prove that the cones which cannot always be avoided are precisely the K-trivial ones. We also prove: (1) If f is convex and X is strongly f-random and Y is Martin-Löf random relative to X, then X is strongly f-random relative to Y. (2) X is complex relative to some oracle if and only if X is random with respect to some continuous probability measure.

MSC

primary
03D32
secondary
68Q30
03D28
03F30

Keywords

Algorithmic randomness
Kolmogorov complexity
Basis theorems
PA-completeness
K-triviality
Partial randomness

Cited by (0)

1

Simpson's research was partially supported by Simons Foundation Collaboration Grant 276282.

2

Stephan's research was partially supported by NUS grant R252-000-420-112.