PPSZ 比你想象的更好
PPSZ is better than you think
AI总结:
本文证明原始 PPSZ 算法在 k-SAT(k≥3)上的性能远超此前认知,通过“假装”变量按精心设计的非均匀分布处理,在不改动算法的情况下将 Unique-3-SAT 的运行时间上界改进到 O(1.306973^n)。
摘要:
PPSZ 长期以来是已知最快的 k-SAT 算法,其工作方式是按照随机顺序遍历输入公式中的变量;随后每个变量被随机设置为 0 或 1,除非可以通过高效可实现的规则(如小宽度归结,或由一小组子句蕴含)推断出正确取值。我们证明,对于所有 k ≥ 3,PPSZ 的表现比先前已知的要好得多,呈指数级改进。对于 Unique-3-SAT,我们将其运行时间上界界定为 O(1.306973^n),这略优于 Hansen、Kaplan、Zamir 和 Zwick 的算法,后者运行时间为 O(1.306995^n)。在此之前,Unique-3-SAT 的已知最佳上界是 O(1.3070319^n)。所有改进都是在不改变原始 PPSZ 的情况下实现的。核心思想是假装 PPSZ 并非按照均匀随机顺序处理变量,而是按照一个精心设计的分布来处理。我们写“假装”是因为这可以在不对算法做任何实际改动的情况下完成。
英文摘要:
PPSZ, for long time the fastest known algorithm for $k$-SAT, works by going through the variables of the input formula in random order; each variable is then set randomly to $0$ or $1$, unless the correct value can be inferred by an efficiently implementable rule (like small-width resolution; or being implied by a small set of clauses). We show that PPSZ performs exponentially better than previously known, for all $k \geq 3$. For Unique-$3$-SAT we bound its running time by $O(1.306973^{n})$, which is somewhat better than the algorithm of Hansen, Kaplan, Zamir, and Zwick, which runs in time $O(1.306995^n)$. Before that, the best known upper bound for Unique-$3$-SAT was $O(1.3070319^n)$. All improvements are achieved without changing the original PPSZ. The core idea is to pretend that PPSZ does not process the variables in uniformly random order, but according to a carefully designed distribution. We write "pretend" since this can be done without any actual change to the algorithm.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。