Skip to main content

VINSE Colloquium Joint Seminar with DSI “Closed-loop design for autonomous materials development” Dr. Kristofer Reyes, University at Buffalo 03/24/2021

Posted by on Wednesday, February 3, 2021 in Events, VINSE Colloquium.

March 24, 2021

Kristofer Reyes
Assistant professor, Department of Materials Design and Innovation
University at Buffalo

“Closed-loop design for autonomous materials development”

4PM via Zoom Webinar
Click HERE to register


Closed-loop, sequential learning is a key paradigm in autonomous materials development. Within this framework, aspects of the materials system under study are modeled, and such models are used to decide subsequent experiments to be run, results of which are fed-back to update models. In the past, off-the-shelf solutions and algorithms have been employed to optimize material properties. In this talk, we describe how autonomous materials platforms can be used in a variety of settings and contexts. We will specifically highlight work in autonomous optimization of additively manufactured mechanical structures, autonomous synthesis of quantum dots by flow chemistry, real-time control of chemical reactions, and closed-loop decision-making in device fabrication. We emphasize how to incorporate physical models and domain-knowledge and expertise to augment experimental data, allow us to both leverage and gain mechanistic physical insight into the material system under study.


Kristofer G. Reyes is an Assistant Professor in the Department of Materials Design and Innovation, University at Buffalo. He applies machine learning and artificial intelligence methods to problems in materials science. He is particularly interested in making these methods relevant in the regime of sparse and noisy data – a regime in which much of materials science research is conducted. A major thrust of his work is in the area of autonomous materials research and development, which uses robot scientists to autonomously plan and execute experiments with little human intervention. In this area, he develops methods for planning optimal experiments based on a limited set of information, algorithms for the autonomous characterization of rich and complex data resulting from experiments, and techniques for autonomously utilizing and learning physics-based knowledge. He received his Ph.D. in Applied Mathematics from the University of Michigan, where he modeled the growth dynamics of nanostructures grown by molecular beam epitaxy and has held positions at Princeton University and the National Security Agency.