Condensed Matter Student Seminar Series



Clark Hall 701


Talk Title: Accelerating Material Experiments with AI-Powered Autonomous Agents for Complex Design Spaces

Abstract: The high dimensionality in the complex modern material design landscape poses challenges in navigating such spaces to find materials of interest. While the parallelization offered by high-throughput experiments helps alleviate this problem, fully exploiting their potential requires effective data analysis workflows. Through the development of SARA, our experiment AI agent, and automated data analysis workflows, we've demonstrated that AI can significantly accelerate certain material experiments. By leveraging the dense physical information from high-throughput data, AI empowers scientists to make decisions at a higher level, drawing upon our physical knowledge and intuition.

(There will be food starting a few minutes before 12 pm.