Optimisation and supervised learning for decision making: competitors or partners?
本文探讨了优化与监督学习在决策过程中的不同角色，概述了二者如何相互促进，并指出了未来研究方向，适合对机器学习与优化交叉领域感兴趣的读者。
Machine learning (ML) is a constantly growing research area. While optimisation researchers are trying to understand how to take advantage of ML methods, ML researchers are developing algorithms to solve classical optimisation problems. Given the huge impact of ML in a large variety of disciplines and application domains, an underlying discussion topic concerns the possibility that ML may become capable of replacing optimisation. As ML is a very broad research area, we focus on supervised learning, one of the most prominent approaches. In this paper we discuss the different roles of optimisation and supervised learning in decision making processes, provide a high level overview of the ways in which each can enrich and benefit the other, and highlight research directions for an impactful interaction of the two paradigms.