The Guessing Secrets problem: a probabilistic approach

We introduce a probabilistic variant of the Guessing Secrets problem proposed by Chung et al. in [Electron. J. Combin. 8 (2001) R13]. In our variation, a player tries to discover the identity of a set S of n unknown secrets drawn by a second player, from a space Ω of N secrets. The first player tries to learn as much as possible about the elements of S by asking binary questions. For each question asked, the second player randomly chooses one of the n secrets of S that he uses in supplying the answer, which in any case must be truthful. We define a simple probabilistic guessing algorithm that allows us to guess all secrets of S with probability one. We show that the expected number of questions needed to guess all secrets is 2n²log₂N and the expected time complexity of the algorithm is . We also propose a generalization of this probabilistic guessing secrets problem, and provide some similar results for this generalization.