Accelerating Dental Adhesive Innovations Through Active Learning and Bayesian Optimization

The discovery of new dental materials is typically a slow process due to high-dimensionality of the formulation space as well as the multiple competing objectives which must be optimized for a given application. Here, we lay out a strategy using active learning and Bayesian optimization that has led to the discovery of 3 new high-performing formulations for dental adhesives within 29 experiments. We utilize curated data from 91 experiments with 43 different components, to reduce the design space and incorporate domain knowledge into our search. The success of this machine learning approach can be adapted to a multitude of dental materials to allow for the fast and efficient discovery of optimal new formulations, leading to enhanced performance, reduced development times, and ultimately more cost-effective and innovative solutions in dental healthcare.