Bayesian Optimization with a Finite Budget: An Approximate Dynamic Programming Approach
Abstract
We consider the problem of optimizing an expensive objective function when a finite budget of total evaluations is prescribed. In that context, the optimal solution strategy for Bayesian optimization can be formulated as a dynamic programming instance. This results in a complex problem with uncountable, dimension-increasing state space and an uncountable control space. We show how to approximate the solution of this dynamic programming problem using rollout, and propose rollout heuristics specifically designed for the Bayesian optimization setting. We present numerical experiments showing that the resulting algorithm for optimization with a finite budget outperforms several popular Bayesian optimization algorithms.
Cite
Text
Lam et al. "Bayesian Optimization with a Finite Budget: An Approximate Dynamic Programming Approach." Neural Information Processing Systems, 2016.Markdown
[Lam et al. "Bayesian Optimization with a Finite Budget: An Approximate Dynamic Programming Approach." Neural Information Processing Systems, 2016.](https://mlanthology.org/neurips/2016/lam2016neurips-bayesian/)BibTeX
@inproceedings{lam2016neurips-bayesian,
title = {{Bayesian Optimization with a Finite Budget: An Approximate Dynamic Programming Approach}},
author = {Lam, Remi and Willcox, Karen and Wolpert, David H.},
booktitle = {Neural Information Processing Systems},
year = {2016},
pages = {883-891},
url = {https://mlanthology.org/neurips/2016/lam2016neurips-bayesian/}
}