BriefGPT.xyz
Jun, 2014
未知高斯过程超参数的贝叶斯优化理论分析
Theoretical Analysis of Bayesian Optimisation with Unknown Gaussian Process Hyper-Parameters
HTML
PDF
Ziyu Wang, Nando de Freitas
TL;DR
在随机环境下,通过使用高斯过程和未知核超参数的贝叶斯优化方法,我们得出了一个对于预期改善收集函数和亚高斯观察噪声的累积遗憾界限,为我们提供了关于如何设计超参数估计方法的指导,并通过简单模拟说明了遵循这些准则的重要性。
Abstract
bayesian optimisation
has gained great popularity as a tool for optimising the parameters of
machine learning algorithms
and models. Somewhat ironically, setting up the hyper-parameters of
→