On Probability of Generating a Finite Group

Let G be a finite group, and let φk(G) be the probability that k random group elements generate G. Denote by θ(G) the smallest k such that φk(G) > 1/e. In this paper we analyze quantity θ(G) for different classes of groups. We prove that θ(G) ≤ κ(G)+ 1 when G is nilpotent and κ(G) is the minimal number of generators of G. When G is solvable we show that θ(G) ≤ 3.25κ(G) + 107. We also show that θ(G) < C log log |G|, where G is a direct product of simple nonabelian groups, and C is a universal constant. The work is motivated by the applications to the “product replacement algorithm” (see [CLMNO,P4]). This algorithm is an important recent innovation, designed to efficiently generate (nearly) uniform random group elements. Recent work by Babai and the author [BaP] showed that the output of the algorithm must have a strong bias in certain cases. The precise probabilistic estimates we obtain here, combined with a work [P3], give a positive result, proving that no bias exists for several families of groups and certain parameters in the algorithm.