We study streaming algorithms for the maximum directed cut problem. The edges of an $n$-vertex directed graph arrive one by one in an arbitrary order, and the goal is to estimate the value of the maximum directed cut using a single pass and small space. With $O(n)$ space, a $(1-\varepsilon)$-approximation can be trivially obtained for any fixed $\varepsilon > 0$ using additive cut sparsifiers. The question that has attracted significant attention in the literature is the best approximation achievable by algorithms that use truly sublinear (i.e., $n^{1-Ω(1)}$) space. A lower bound of Kapralov and Krachun (STOC'20) implies .5-approximation is the best one can hope for. The current best algorithm for general graphs obtains a .485-approximation due to the work of Saxena, Singer, Sudan, and Velusamy (FOCS'23). The same authors later obtained a $(1/2-\varepsilon)$-approximation, assuming that the graph is constant-degree (SODA'25). In this paper, we show that for any $\varepsilon > 0$, a $(1/2-\varepsilon)$-approximation of maximum dicut value can be obtained with $n^{1-Ω_\varepsilon(1)}$ space in *general graphs*. This shows that the lower bound of Kapralov and Krachun is generally tight, settling the approximation complexity of this fundamental problem. The key to our result is a careful analysis of how correlation propagates among high- and low-degree vertices, when simulating a suitable local algorithm.
翻译:本文研究最大有向割问题的流式算法。给定一个包含 $n$ 个顶点的有向图,其边以任意顺序逐一到达,目标是通过单遍扫描和少量存储空间来估计最大有向割的值。对于任意固定的 $\varepsilon > 0$,利用加法割稀疏化技术,可在 $O(n)$ 空间内平凡地获得 $(1-\varepsilon)$ 近似解。文献中备受关注的核心问题是:在使用真正亚线性(即 $n^{1-Ω(1)}$)空间的算法中,可达到的最佳近似比是多少。Kapralov 与 Krachun(STOC'20)的下界表明 .5 近似是可能达到的最优结果。目前针对一般图的最佳算法由 Saxena、Singer、Sudan 和 Velusamy(FOCS'23)提出,实现了 .485 近似。同一组作者后续在假设图为常度数的条件下(SODA'25)得到了 $(1/2-\varepsilon)$ 近似。本文证明,对于任意 $\varepsilon > 0$,在*一般图*中可通过 $n^{1-Ω_\varepsilon(1)}$ 空间获得最大有向割值的 $(1/2-\varepsilon)$ 近似。这表明 Kapralov 与 Krachun 的下界在一般情况下是紧的,从而解决了这一基础问题的近似计算复杂度。我们结果的关键在于,当模拟合适的局部算法时,对高、低度数顶点间相关性传播机制的精细分析。