Enhanced Bayesian Optimization via Preferential Modeling of Abstract Properties

TL;DR

This paper introduces a collaborative Bayesian Optimization framework that integrates human expert preferences on abstract properties to improve optimization efficiency, robustness, and convergence in experimental design tasks.

Contribution

It presents a novel method for incorporating expert preferences on unmeasured properties into Bayesian Optimization, enhancing performance and robustness against biased judgments.

Findings

Outperforms baseline methods on synthetic and real-world datasets.

Effectively handles incorrect or misleading expert biases.

Demonstrates improved convergence behavior.

Abstract

Experimental (design) optimization is a key driver in designing and discovering new products and processes. Bayesian Optimization (BO) is an effective tool for optimizing expensive and black-box experimental design processes. While Bayesian optimization is a principled data-driven approach to experimental optimization, it learns everything from scratch and could greatly benefit from the expertise of its human (domain) experts who often reason about systems at different abstraction levels using physical properties that are not necessarily directly measured (or measurable). In this paper, we propose a human-AI collaborative Bayesian framework to incorporate expert preferences about unmeasured abstract properties into the surrogate modeling to further boost the performance of BO. We provide an efficient strategy that can also handle any incorrect/misleading expert bias in preferential judgments. We discuss the convergence behavior of our proposed framework. Our experimental results involving synthetic functions and real-world datasets show the superiority of our method against the baselines.