Deixei
(onde é considerado codificado em binário). Então, o que podemos dizer sobre a complexidade computacional de L ? É claro que L ∈ E X P . E, se não me engano, os incríveis algoritmos "do tipo BBP" para calcular o n t h bit de π usando tempo quase linear e ( log n ) O ( 1 ) de memória, sem precisar calcular os bits anteriores, produzem L ∈ P S P A C E .
Podemos melhorar ainda mais e colocar (por exemplo) na hierarquia de contagem? Na outra direção, existe algum resultado de dureza para L (mesmo um extremamente fraco, como a dureza T C 0 )?
Uma linguagem relacionada interessante é
(onde, novamente, é escrito em binário). Nós temos
e, portanto, ; Eu ficaria extremamente interessado se algo melhor fosse conhecido.
cc.complexity-theory
na.numerical-analysis
Scott Aaronson
fonte
fonte
Respostas:
OK, James Lee has pointed me to this 2011 paper by Samir Datta and Rameshwar Pratap, which proves that my languageL (encoding the digits of π ) is in the fourth level of the counting hierarchy (PHPPPPPP ; thanks to SamiD below for pointing out a missing PP in the paper, which I'd simply repeated in my answer!). The paper also explicitly discusses my question of lower bounds on the complexity of computing the binary digits of irrational numbers, though it only manages to prove a very weak lower bound for computing the binary digits of rational numbers. This is exactly what I was looking for.
Update (April 3): An amusing consequence of the digits ofπ being computable in the counting hierarchy is as follows. Suppose that π is a normal number (whose binary expansion converges quickly to "effectively random"), and suppose that P=PP (with the simulation involving only a small polynomial overhead). Then it would be feasible to program your computer to find, for example, the first occurrence of the complete works of Shakespeare in the binary expansion of π . If that sounds absurd to you, then maybe it should be taken as additional evidence that P≠PP . :-)
fonte