Santini

le 7 Décembre 2000 à 14h30

au LRI, Salle 101

M. Santini

(Université de Milan)

On the sequential and circuit complexity of the random generation problem for polynomially ambiguous context-free languages

Résumé/Abstract :

We study the sequential and circuit complexity of generating at random, under uniform distribution, a word of length n from a given context-free language. We show that for every inherently finitely ambiguous context-free language the problem can be solved in O(n2 log n) time on a RAM under logarithmic cost criterion; moreover we show that for every inherently polynomially ambiguous context-free language the problem can be solved by a logspace-uniform family of probabilistic boolean circuits of polynomial size and O(log2 n) depth. Using a suitable notion of reducibility (similar to the NC1-reducibility), we also show the relationship between random generation problems and classical computational complexity classes such as DIV, L and DET.