English

Average-Case Subset Balancing Problems

Computational Complexity 2021-10-28 v1 Data Structures and Algorithms

Abstract

Given a set of nn input integers, the Equal Subset Sum problem asks us to find two distinct subsets with the same sum. In this paper we present an algorithm that runs in time O(30.387n)O^*(3^{0.387n}) in the~average case, significantly improving over the O(30.488n)O^*(3^{0.488n}) running time of the best known worst-case algorithm and the Meet-in-the-Middle benchmark of O(30.5n)O^*(3^{0.5n}). Our algorithm generalizes to a number of related problems, such as the ``Generalized Equal Subset Sum'' problem, which asks us to assign a coefficient cic_i from a set CC to each input number xix_i such that icixi=0\sum_{i} c_i x_i = 0. Our algorithm for the average-case version of this problem runs in~time C(0.5c0/C)n|C|^{(0.5-c_0/|C|)n} for some positive constant c0c_0, whenever C={0,±1,,±d}C=\{0, \pm 1, \dots, \pm d\} or {±1,,±d}\{\pm 1, \dots, \pm d\} for some positive integer dd (with O(C0.45n)O^*(|C|^{0.45n}) when C<10|C|<10). Our results extend to the~problem of finding ``nearly balanced'' solutions in which the target is a not-too-large nonzero offset τ\tau. Our approach relies on new structural results that characterize the probability that icixi\sum_{i} c_i x_i =τ=\tau has a solution cCnc \in C^n when xix_i's are chosen randomly; these results may be of independent interest. Our algorithm is inspired by the ``representation technique'' introduced by Howgrave-Graham and Joux. This requires several new ideas to overcome preprocessing hurdles that arise in the representation framework, as well as a novel application of dynamic programming in the solution recovery phase of the algorithm.

Keywords

Cite

@article{arxiv.2110.14607,
  title  = {Average-Case Subset Balancing Problems},
  author = {Xi Chen and Yaonan Jin and Tim Randolph and Rocco A. Servedio},
  journal= {arXiv preprint arXiv:2110.14607},
  year   = {2021}
}

Comments

44 pages, 5 figures