arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1904.09354cs.DScs.LGmath.OC
收藏

超越非负性的次模最大化:保证、快速算法及应用

Submodular Maximization Beyond Non-negativity: Guarantees, Fast Algorithms, and Applications

Yale University(耶鲁大学);Open University of Israel(以色列开放大学);Queen Mary University of London(伦敦玛丽女王大学)

Christopher Harshaw, Moran Feldman, Justin Ward, Amin Karbasi

摘要:

一般认为,次模函数--以及更一般的$γ$类--弱次模函数--只能在非负性假设$f(S) \geq 0$下优化。在本文中,我们表明,一旦函数表示为差$f = g - c$,其中$g$是单调的,非负的,和$γ$-弱次模和$c$是非负模,那么强逼近保证可能会得到。我们提出了一个在$k$基数约束下最大化$g - c$的算法,该算法产生一个随机可行集$S$,使得运行时间为$O (\frac{n}ε \log^2 \frac{1}ε)$的$\mathbb{E} \left[ g(S) - c(S) \right] \geq (1 - e^{-γ} - ε) g(OPT) - c(OPT)$,即,独立于$k$。我们将这些结果扩展到无约束的设置,通过描述具有相同的近似保证和更快的$O(\frac{n}ε \log\frac{1}ε)$运行时的算法。我们的算法的主要技术是双重的:使用一个代理目标,不同的$g$和$c$之间的相对重要性在整个算法,和几何扫描可能的$γ$值。我们的算法保证补充了硬度的结果表明,没有多项式时间的算法访问$g$通过一个值Oracle可以做得更好。我们经验证明我们的算法的成功,将它们应用到波士顿住房数据集的实验设计和有向顶点覆盖的电子邮件欧盟数据集。

英文摘要:

It is generally believed that submodular functions -- and the more general class of $γ$-weakly submodular functions -- may only be optimized under the non-negativity assumption $f(S) \geq 0$. In this paper, we show that once the function is expressed as the difference $f = g - c$, where $g$ is monotone, non-negative, and $γ$-weakly submodular and $c$ is non-negative modular, then strong approximation guarantees may be obtained. We present an algorithm for maximizing $g - c$ under a $k$-cardinality constraint which produces a random feasible set $S$ such that $\mathbb{E} \left[ g(S) - c(S) \right] \geq (1 - e^{-γ} - ε) g(OPT) - c(OPT)$, whose running time is $O (\frac{n}ε \log^2 \frac{1}ε)$, i.e., independent of $k$. We extend these results to the unconstrained setting by describing an algorithm with the same approximation guarantees and faster $O(\frac{n}ε \log\frac{1}ε)$ runtime. The main techniques underlying our algorithms are two-fold: the use of a surrogate objective which varies the relative importance between $g$ and $c$ throughout the algorithm, and a geometric sweep over possible $γ$ values. Our algorithmic guarantees are complemented by a hardness result showing that no polynomial-time algorithm which accesses $g$ through a value oracle can do better. We empirically demonstrate the success of our algorithms by applying them to experimental design on the Boston Housing dataset and directed vertex cover on the Email EU dataset.

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。

↑