A nearly optimal randomized algorithm for explorable heap selection

TL;DR

A new randomized algorithm is presented for selecting the nth smallest value in a binary heap, improving efficiency while using more space.

Contribution

A nearly optimal randomized algorithm for heap selection with improved runtime and a matching lower bound.

Findings

The new algorithm runs in O(n log(n)^3) time with O(log n) space.

A lower bound of Ω(log(n)n / log(log(n))) is established for the problem.

The algorithm improves upon previous randomized methods at the cost of slightly increased space.

Abstract

Explorable heap selection is the problem of selecting the nth smallest value in a binary heap. The key values can only be accessed by traversing through the underlying infinite binary tree, and the complexity of the algorithm is measured by the total distance traveled in the tree (each edge has unit cost). This problem was originally proposed as a model to study search strategies for the branch-and-bound algorithm with storage restrictions by Karp, Saks and Widgerson (FOCS ’86), who gave deterministic and randomized \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$n\cdot \exp (O(\sqrt{\log {n}}))$$\end{document}n·exp(O(logn)) time algorithms using \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$O(\log (n)^{2.5})$$\end{document}O(log(n)2.5) and \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$O(\sqrt{\log n})$$\end{document}O(logn) space respectively. We present a new randomized algorithm with running time \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$O(n\log (n)^3)$$\end{document}O(nlog(n)3) against an oblivious adversary using \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$O(\log n)$$\end{document}O(logn) space, substantially improving the previous best randomized running time at the expense of slightly increased space usage. We also show an \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\Omega (\log (n)n/\log (\log (n)))$$\end{document}Ω(log(n)n/log(log(n))) lower bound for any algorithm that solves the problem in the same amount of space, indicating that our algorithm is nearly optimal.

Figures