DT-PBO: an Interpretable Tree-based Surrogate Model for Preferential Bayesian Optimization

TL;DR

DT-PBO introduces an interpretable, tree-based surrogate model for preferential Bayesian optimization, effectively balancing preference modeling accuracy with transparency, especially in high-stakes applications.

Contribution

The paper presents a novel shallow decision tree surrogate with a new splitting heuristic for PBO, enhancing interpretability without sacrificing performance.

Findings

Achieves competitive convergence with GP-based PBO on benchmark functions.

Demonstrates robustness to noise and fast computation.

Provides interpretable insights into preferences in real-world data.

Abstract

Preferential Bayesian Optimization (PBO) aims to find a decision-maker's most preferred solution in as few pairwise comparisons as possible. Existing approaches rely on Gaussian Process (GP) surrogates, which provide strong performance but limited interpretability. This limits real-world usability in high-stakes domains, such as healthcare, where interpretability and trust are essential. We propose DT-PBO, a novel tree-based surrogate model for PBO that is inherently interpretable while capturing preference uncertainty. Specifically, we introduce a novel splitting heuristic that constructs interpretable shallow decision trees directly from pairwise comparison data, and use Laplace approximation to obtain probabilistic estimates for each leaf. This enables efficient preference modeling without sacrificing interpretability. Across eight benchmark functions, our method achieves competitive convergence to GP-based PBO, particularly on functions with rugged optimization landscapes. Additional experiments show robustness against noise and a fast computational running time. Experiments on real-world datasets further demonstrate that our model provides interpretable insights into decision-maker preferences that would remain opaque under GP-based approaches.