Human-in-the-loop for Bayesian autonomous materials phase mapping

Published

February 7, 2024

Author(s)

Felix Adams, Austin McDannald, Ichrio Takeuchi, A. Gilad Kusne

Abstract

Autonomous experimentation achieves user objectives more efficiently than Edisonian studies by combining machine learning and laboratory automation to iteratively select and perform experiments. Integrating knowledge from theory, simulations, literature, and human intuition into the machine learning model can further increase this advantage. We present a set of methods for probabilistically integrating human input into an autonomous materials exploration campaign for composition-structure phase mapping. During the campaign, the user can provide input by indicating potential phase boundaries or phase regions with their uncertainty or indicating regions of interest. The input is then integrated through probabilistic priors, resulting in a probabilistic distribution over potential phase maps given the data, model, and human input. We demonstrate an improvement in phase-mapping performance given appropriate human input.

Citation

Matter

Volume

7

Issue

2

Pub Type

Journals

Download Paper

https://doi.org/10.1016/j.matt.2024.01.005

Local Download

Citation

Adams, F. , McDannald, A. , Takeuchi, I. and Kusne, A. (2024), Human-in-the-loop for Bayesian autonomous materials phase mapping, Matter, [online], https://doi.org/10.1016/j.matt.2024.01.005, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=956132 (Accessed October 13, 2025)