Work

Thermodynamically informed priors for uncertainty propagation in first-principles statistical mechanics

Year: 2024

Type: article

Abstract: This work demonstrates how first-principles statistical mechanics approaches within a Bayesian framework can quantify and propagate uncertainties to downstream thermodynamic calculations. To address the issue of Bayesian prior selection, knowledge of 0 K ground states in the material system of interest is incorporated into the prior. The effectiveness of this framework is shown by creating a phase diagram for the fcc zirconium nitride system, including confidence intervals on order-disorder transition temperatures.

Source: Physical Review Materials

DOI: 10.1103/physrevmaterials.8.103803

Authors: Derick E. Ober, Anton Van der Ven

Institution: University of California, Santa Barbara

Language: English

Topic: Machine Learning in Materials Science

Subfield: Materials Chemistry

Field: Materials Science

Domain: Physical Sciences

Open Access status: green

Funders: National Science Foundation, U.S. Department of Energy, National Energy Research Scientific Computing Center, Office of Science

Awards: G1110399006, SOFTWARE CENTER FOR PREDICTIVE THEORY AND MODELING, Materials Research Science and Engineering Center at UCSB, G419881185, G8920200208