Bayesian Optimization Meets Search Based Optimization: A Hybrid Approach for Multi-Fidelity Optimization
Bayesian Optimization Meets Search Based Optimization: A Hybrid Approach for Multi-Fidelity Optimization
复制标题
贝叶斯优化与基于搜索的优化的结合：多保真度优化的混合方法
DOI：
10.1609/aaai.v32i1.12184
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
J. Doppa
中科院分区：
文献类型：
--
作者：
Ellis Hoag;J. Doppa
许多现实生活中的问题都需要用昂贵的评估来优化函数。贝叶斯优化(BO)和基于搜索的优化(SO)是两大类算法，它们试图寻找函数的全局最优值，目标是最小化函数求值的次数。现有的大量工作涉及单一保真度设置，其中函数评估非常昂贵但准确。然而，在许多应用中，我们可以访问多保真函数，这些函数的成本和评估精度各不相同。在本文中，我们提出了一种新的方法，称为多保真度混合(多保真混合)，它结合了BO方法和SO方法的最佳属性，以最小的代价发现黑盒函数的全局最优解。在多个基准函数上的实验表明，该算法的性能优于已有的单保真和多保真优化算法。
Many real-life problems require optimizing functions with expensive evaluations. Bayesian Optimization (BO) and Search-based Optimization (SO) are two broad families of algorithms that try to find the global optima of a function with the goal of minimizing the number of function evaluations. A large body of existing work deals with the single-fidelity setting, where function evaluations are very expensive but accurate. However, in many applications, we have access to multiple-fidelity functions that vary in their cost and accuracy of evaluation. In this paper, we propose a novel approach called Multi-fidelity Hybrid (MF-Hybrid) that combines the best attributes of both BO and SO methods to discover the global optima of a black-box function with minimal cost. Our experiments on multiple benchmark functions show that the MF-Hybrid algorithm outperforms existing single-fidelity and multi-fidelity optimization algorithms.