arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.04769math.OCcs.DScs.LG
收藏

次模块化+凹面

Submodular + Concave

Yale University(耶鲁大学);University of Haifa(海法大学)

Siddharth Mitra, Moran Feldman, Amin Karbasi

摘要:

一阶最优化方法可以收敛到凹函数的最大目标值,并为(非凸/非凹)连续次模函数提供常数因子逼近保证。在这项工作中,我们开始研究的形式$F(x) = G(x) +C(x)$在可解凸体$P$,其中$G$是一个光滑的DR-次模函数和$C$是一个光滑的凹函数的函数的最大化。这类函数是凹的和连续的DR-次模函数的严格扩展,对于这些次模函数没有理论保证。我们提供了一套Frank-Wolfe风格的算法,根据目标函数的性质(即,如果$G$和$C$是单调的或非单调的,以及非负的或非负的)以及集合$P$的性质(即,无论是否向下闭合),提供$1-1/e$、$1/e$或$1/2$近似保证。然后,我们使用我们的算法得到一个框架,从一个给定的地面集(对应于一个决定性的点过程的模式),并选择一组聚集的元素(对应于一个合适的凹函数的最大值)选择不同的元素之间平滑插值。此外,我们将我们的算法应用于上述类(DR-次模+凹)中的各种函数,在约束和无约束的设置,并表明我们的算法始终优于自然基线。

英文摘要:

It has been well established that first order optimization methods can converge to the maximal objective value of concave functions and provide constant factor approximation guarantees for (non-convex/non-concave) continuous submodular functions. In this work, we initiate the study of the maximization of functions of the form $F(x) = G(x) +C(x)$ over a solvable convex body $P$, where $G$ is a smooth DR-submodular function and $C$ is a smooth concave function. This class of functions is a strict extension of both concave and continuous DR-submodular functions for which no theoretical guarantee is known. We provide a suite of Frank-Wolfe style algorithms, which, depending on the nature of the objective function (i.e., if $G$ and $C$ are monotone or not, and non-negative or not) and on the nature of the set $P$ (i.e., whether it is downward closed or not), provide $1-1/e$, $1/e$, or $1/2$ approximation guarantees. We then use our algorithms to get a framework to smoothly interpolate between choosing a diverse set of elements from a given ground set (corresponding to the mode of a determinantal point process) and choosing a clustered set of elements (corresponding to the maxima of a suitable concave function). Additionally, we apply our algorithms to various functions in the above class (DR-submodular + concave) in both constrained and unconstrained settings, and show that our algorithms consistently outperform natural baselines.

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。

↑