Year: 2022
Type: article
Abstract: Data-driven design shows the promise of accelerating materials discovery but is challenging due to the prohibitive cost of searching the vast design space of chemistry, structure, and synthesis methods. Bayesian optimization (BO) employs uncertainty-aware machine learning models to select promising designs to evaluate, hence reducing the cost. However, BO with mixed numerical and categorical variables, which is of particular interest in materials design, has not been well studied. In this work, we survey frequentist and Bayesian approaches to uncertainty quantification of machine learning with mixed variables. We then conduct a systematic comparative study of their performances in BO using a popular representative model from each group, the random forest-based Lolo model (frequentist) and the latent variable Gaussian process model (Bayesian). We examine the efficacy of the two models in the optimization of mathematical functions, as well as properties of structural and functional materials, where we observe performance differences as related to problem dimensionality and complexity. By investigating the machine learning models' predictive and uncertainty estimation capabilities, we provide interpretations of the observed performance differences. Our results provide practical guidance on choosing between frequentist and Bayesian uncertainty-aware machine learning models for mixed-variable BO in materials design.
Source: Scientific Reports
Institution: Northwestern University
Language: English
Subfield: Materials Chemistry
Field: Materials Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,190
Awards: G2703334496, Collaborative Research: Dynamics of Intraplate Fault Systems, Northern Basin and Range Province, FMRG: Manufacturing ADvanced Electronics through Printing Using Bio-based and Locally Identifiable Compounds (MADE-PUBLIC), THIS COOPERATIVE AGREEMENT IS AWARDED TO NORTHWESTERN UNIVERSITY FOR A PROJECT ENTITLED ''ADAPTIVE DISCOVERY AND MIXED-VARIABLE OPTIMIZATION OF NEXT GENERATION SYNTHESIZABLE MICROELECTRONIC MATERIALS”.
THIS AWARD IS BEING MADE TO YOUR ORGANIZATION IN ACCORDANCE WITH FOA NUMBER DE-FOA-0002107 (DIFFERENTIATE) AND APPLICATION CONTROL NUMBER 2107-1543.
THE FOLLOWING ATTACHMENTS ARE INCLUDED IN THIS AWARD:
ATTACHMENT 1: SPECIAL TERMS AND CONDITIONS
ATTACHMENT 2: INTELLECTUAL PROPERTY PROVISIONS
ATTACHMENT 3: TECHNICAL MILESTONES AND DELIVERABLES
ATTACHMENT 4: ARPA-E REPORTING CHECKLIST AND INSTRUCTIONS
ATTACHMENT 5: BUDGET INFORMATION
ATTACHMENT 6: NATIONAL POLICY ASSURANCES
NOTE TO BLOCK 12 OF PAGE 1 (TOTAL AMOUNT):
THE AMOUNT OBLIGATED BY ARPA-E TO THIS COOPERATIVE AGREEMENT IS $1,521,898.00. THE PRIME RECIPIENT WILL PROVIDE $362,000 IN COST SHARE CONTRIBUTIONS.
THE TOTAL COOPERATIVE AGREEMENT AMOUNT IS $1,624,513.00.
BY ACCEPTING THIS AGREEMENT, NORTHWESTERN UNIVERSITY AFFIRMS THAT THEY HAVE RECEIVED NO OTHER FEDERAL FUNDING FOR THIS PROJECT.
PLEASE INDICATE THE AWARD NUMBER AR-DE0001209 ON ALL DOCUMENTS SUBMITTED.
OpenAlex ID: W4309403833