Range-Aware Bayesian Optimization for Discovering Diverse Designs within Target Property Windows

2026-06-09

AI Summary

This work addresses the challenge in materials and product design where the objective is often to discover diverse solutions that satisfy a target performance interval—rather than a single optimum—while accounting for hard-to-quantify factors such as cost and manufacturability. The authors propose an interval-aware Bayesian optimization framework that, for the first time, directly incorporates performance intervals into the acquisition function by evaluating the posterior probability that a candidate design meets the desired specifications. Combining Gaussian process modeling, a probabilistic acquisition function under interval constraints, and a parallel multi-specification exploration strategy, the method enables efficient and diverse specification-driven design. Empirical results on real-world tasks—including polymer synthesis and oligomer optical bandgap design—demonstrate clear superiority over existing approaches, yielding richer sets of valid, specification-compliant designs.

Abstract

In many materials and product design problems, desirable candidates exhibit properties that fall within an acceptable range rather than achieve a single optimum. Recovering multiple, distinct solutions that satisfy such specifications is also practically valuable, as some candidates may be preferred for reasons of cost, processability, or robustness that are difficult to encode directly in an objective function. Here, we develop a range-aware Bayesian optimization (BO) framework in which the acquisition function directly scores the posterior probability that a candidate satisfies a target range. The framework naturally extends to parallel pursuit of multiple distinct specifications over a shared candidate space. Across benchmark tasks, range-aware acquisition consistently recovers larger and more diverse sets of valid designs than standard BO baselines and recent goal-seeking methods. Its utility is further demonstrated in two practically motivated design case studies involving optimizing reaction conditions for polymer synthesis and sequence-defined oligomer discovery for prescribed optical absorption bands, supported by quantum chemical calculations. These results suggest that range-aware BO can provide a practical and sample-efficient foundation for specification-driven design, particularly when design flexibility and solution diversity are important considerations.

Problem

• range-aware
• Bayesian optimization
• diverse designs
• target property windows
• specification-driven design

Innovation

• range-aware Bayesian optimization
• target property windows
• diverse design discovery
• acquisition function
• specification-driven design