About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
AI/Data Informatics: Computational Model Development, Verification, Validation, and Uncertainty Quantification
|
Presentation Title |
Experimentally Validated High-dimensional Bayesian Optimization of Dental Adhesives via Adaptive Design |
Author(s) |
Ramsey Issa, Taylor D. Sparks |
On-Site Speaker (Planned) |
Ramsey Issa |
Abstract Scope |
Although dental adhesives play a crucial role in overall oral health, they are yet to feel the impact of Bayesian optimization guided adaptive design strategies. This results in a slow development process for these crucial adhesives. Typically, dental adhesives are high-dimensional materials that contain resins, fillers, inhibitors, initiators, solvents, acids, and antimicrobial agents. The choice of features leads to design spaces of thirty different possible inputs into a single adhesive. This leads to an exhaustive design space that is difficult to tackle using strictly domain knowledge. Here, we employ the sparse axis-aligned subspace Bayesian optimization (SAASBO) method to optimize the shear bonding to dentin. We experimentally validate through synthesis and characterization that using SAASBO in high-dimensional materials adaptive design strategies leads to the discovery of new materials. |
Proceedings Inclusion? |
Planned: |
Keywords |
Machine Learning, Computational Materials Science & Engineering, Modeling and Simulation |