We are developing machine learning-driven autonomous research systems, designed with the goal of accelerating the discovery and optimization of advanced materials. These systems combine machine learning with machine-controlled materials synthesis and characterization tools for closed loop experiment design, execution, and analysis, where experiment design is guided by active learning, Bayesian optimization, and similar methods. A key challenge is the integration of prior physics knowledge into the data analysis, including both physics theory and databases of experimental and computational materials property data.

We are particularly interested in using these autonomous systems to verify and identify phase maps for thin films, bulk, and surface morphology of solid state materials as well as phase maps for aqueous electrochemical materials.

The team also has ongoing projects for offline and on-the-fly (e.g., real-time, during the measurement experiment) analysis of high throughput combinatorial “library” experiments for a diverse set of materials applications. We are particularly interested in hyperspectral methods (algorithms for analyzing spectra collected over 2D coordinates) to analyze X-ray diffraction and Raman spectra data collected from combinatorial libraries as well as hyperspectral micrographs. Some examples of materials classes of interest for this project are metallic glasses, photovoltaic, superconductive, multiferroic, thermoelectric, thermochromic, and magnetic materials.

key words

Materials Genome Initiative; Autonomous; Machine learning; Informatics; High-throughput; Data mining; Functional materials; Active Learning

Citizenship:  Open to U.S. citizens

Level:  Open to Postdoctoral applicants