On Uncertainty Estimation by Tree-based Surrogate Models in Sequential Model-based Optimization

TL;DR

This paper investigates the effectiveness of ensemble tree-based models for uncertainty estimation in sequential model-based optimization, proposing a new BwO forest method that outperforms existing models in various scenarios.

Contribution

It introduces BwO forest, a novel ensemble of randomized trees with oversampling, enhancing uncertainty estimation for optimization tasks.

Findings

BwO forest demonstrates superior uncertainty estimation accuracy.

The proposed method outperforms existing tree-based models.

Experimental results validate the effectiveness of BwO forest.

Abstract

Sequential model-based optimization sequentially selects a candidate point by constructing a surrogate model with the history of evaluations, to solve a black-box optimization problem. Gaussian process (GP) regression is a popular choice as a surrogate model, because of its capability of calculating prediction uncertainty analytically. On the other hand, an ensemble of randomized trees is another option and has practical merits over GPs due to its scalability and easiness of handling continuous/discrete mixed variables. In this paper we revisit various ensembles of randomized trees to investigate their behavior in the perspective of prediction uncertainty estimation. Then, we propose a new way of constructing an ensemble of randomized trees, referred to as BwO forest, where bagging with oversampling is employed to construct bootstrapped samples that are used to build randomized trees with random splitting. Experimental results demonstrate the validity and good performance of BwO forest over existing tree-based models in various circumstances.