arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1008.2159cs.DScs.DMcs.LG
收藏

次模函数:可学习性、结构与优化

Submodular Functions: Learnability, Structure, and Optimization

Georgia Institute of Technology(佐治亚理工学院);University of British Columbia(不列颠哥伦比亚大学)

Maria-Florina Balcan, Nicholas J. A. Harvey

摘要:

次模函数是离散函数,它模拟收益递减规律,并享有许多算法应用。它们已被用于许多领域,包括组合优化、机器学习和经济学。本文从学习理论的角度研究了次模函数。我们提供学习次模函数的算法,以及它们的可学习性的下界。在这样做的时候,我们发现了一些新的结构结果揭示了子模块功能可以令人惊讶的结构化和令人惊讶的非结构化的方式。我们提供了一些具体的影响,我们的工作在其他领域,包括算法博弈论和组合优化。在技术层面上,这项研究结合了许多领域的思想,包括学习理论(分布式学习和PAC风格的分析),组合学和优化(拟阵和子模函数),以及伪随机性(无损扩展图)。

英文摘要:

Submodular functions are discrete functions that model laws of diminishing returns and enjoy numerous algorithmic applications. They have been used in many areas, including combinatorial optimization, machine learning, and economics. In this work we study submodular functions from a learning theoretic angle. We provide algorithms for learning submodular functions, as well as lower bounds on their learnability. In doing so, we uncover several novel structural results revealing ways in which submodular functions can be both surprisingly structured and surprisingly unstructured. We provide several concrete implications of our work in other domains including algorithmic game theory and combinatorial optimization. At a technical level, this research combines ideas from many areas, including learning theory (distributional learning and PAC-style analyses), combinatorics and optimization (matroids and submodular functions), and pseudorandomness (lossless expander graphs).

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。

↑