Fast Approximate Rank Determination and Selection with Group Testing

Abstract: Suppose that a group test operation is available for checking order relations in a set, can this speed up problems like finding the minimum/maximum element, rank determination and selection? We consider a one-sided group test to be available, where queries are of the form $u \le_Q V$ or $V \le_Q u$, and the answer is `yes' if and only if there is some $v \in V$ such that $u \le v$ or $v \le u$, respectively. We restrict attention to total orders and focus on query-complexity; for min or max finding, we give a Las Vegas algorithm that makes $\mathcal{O}(\log^2 n)$ expected queries. We also give randomized approximate algorithms for rank determination and selection; we allow a relative error of $1 \pm \delta$ for $\delta > 0$ in the estimated rank or selected element. In this case, we give a Monte Carlo algorithm for approximate rank determination with expected query complexity $\tilde{\mathcal{O}}(1/\delta^2 - \log \epsilon)$, where $1-\epsilon$ is the probability that the algorithm succeeds. We also give a Monte Carlo algorithm for approximate selection that has expected query complexity $\tilde{\mathcal{O}}(-\log( \epsilon \delta^2) / \delta^4)$; it has probability at least $\frac{1}{2}$ to output an element $x$, and if so, $x$ has the desired approximate rank with probability $1-\epsilon$.