Scholarlink

This chapter presents the main contribution of the authors of this book, that is, a multiple information source Bayesian optimization algorithm based on a novel Gaussian process model for this specific setting. A Python implementation of the approach is presented, along with experiments on a set of test problems. The proposed method is compared against other multi-fidelity Bayesian optimization methods, demonstrating its main advantages.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

MISO-AGP: Dealing with Multiple Information Sources via Augmented Gaussian Process

  • 章节

  • 发表日期：

    2025年

查看全文
  • Antonio Candelieri,
  • Andrea Ponti,
  • Francesco Archetti

摘要

This chapter presents the main contribution of the authors of this book, that is, a multiple information source Bayesian optimization algorithm based on a novel Gaussian process model for this specific setting. A Python implementation of the approach is presented, along with experiments on a set of test problems. The proposed method is compared against other multi-fidelity Bayesian optimization methods, demonstrating its main advantages.