论文标题
Stepdirect-一种逐步函数的无衍生化优化方法
StepDIRECT -- A Derivative-Free Optimization Method for Stepwise Functions
论文作者
论文摘要
在本文中，我们提出了用于无衍生优化（DFO）的级别算法，其中黑盒目标函数具有逐步的景观。我们的框架基于众所周知的直接算法。通过合并局部变异性来探索平坦度，我们提供了一个新的标准，以选择潜在的最佳超矩形。此外，我们引入了一种随机局部搜索算法，该算法在潜在的最佳超矩形上执行，以提高溶液质量和收敛速度。提供了继总算法的全局收敛。提出了有关对随机森林模型和高参数调整优化的数值实验，以支持我们的算法的功效。与其他最先进的基线DFO方法相比，所提出的继承算法显示出竞争性能结果，包括原始直接算法。
In this paper, we propose the StepDIRECT algorithm for derivative-free optimization (DFO), in which the black-box objective function has a stepwise landscape. Our framework is based on the well-known DIRECT algorithm. By incorporating the local variability to explore the flatness, we provide a new criterion to select the potentially optimal hyper-rectangles. In addition, we introduce a stochastic local search algorithm performing on potentially optimal hyper-rectangles to improve the solution quality and convergence speed. Global convergence of the StepDIRECT algorithm is provided. Numerical experiments on optimization for random forest models and hyper-parameter tuning are presented to support the efficacy of our algorithm. The proposed StepDIRECT algorithm shows competitive performance results compared with other state-of-the-art baseline DFO methods including the original DIRECT algorithm.
