Neural Meta Learning for Multifidelity Bayesian Optimization

碩士 === 國立交通大學 === 電機工程學系 === 107 === Bayesian optimization (BO) has recently demonstrated to be powerful in optimizing an unknown (possibly non-convex with no closed-form derivative) target function using a finite budget to evaluate a complicated and computationally expensive function. BO methods ty...

Full description

Bibliographic Details
Main Authors: Lieow, Wei Xiang, 廖偉翔
Other Authors: Chien, Jen-Tzung
Format: Others
Language:en_US
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/7wz9m9