Materials Science Collaborators Converge at Symposium About AI in the Field

In early September, more than 70 scientists, engineers, trainees, and industry experts gathered at Duke University’s Trent Semans Center for Health Education for the Harnessing Data for Materials Symposium. The two-day event brought together participants from seven leading universities, five of which host a National Research Trainee (NRT) program focused on integrating machine learning with materials science. The event was hosted by Duke’s AI for Understanding and Designing Materials (aiM) NRT program and led by Duke faculty members Shana McAlexander and Cate Brinson.  

The symposium was designed to promote cross-disciplinary collaboration and to respond to the National Science Foundation’s Harnessing the Data Revolution (HDR) initiative, which aims to accelerate advancements in materials science through the application of AI and data-driven methods.

The symposium served as a platform for trainees, faculty, and industry professionals to share research knowledge, explore new research opportunities, and strengthen collaborations at the intersection of artificial intelligence and materials science. Over the course of the event, attendees engaged in formal presentations, roundtable discussions, and team-based activities that emphasized the power of interdisciplinary teamwork. 

For many participants, the symposium was a chance to not only deepen their understanding of materials science but also to explore how AI could help overcome some of the field’s most pressing challenges.

“I gained fresh insights into the application of specific machine learning algorithms in material sciences, particularly around how common research models lack sufficient data,” said Yiyang Sun, a computer science PhD candidate at Duke. “The roundtables and lightning talks provided excellent opportunities to exchange perspectives and learn from others.”

One of highlights of the symposium was the keynote address delivered by Dr. Olga Wodo, an associate professor at the University at Buffalo. Wodo’s talk, titled Microstructure Informatics: Importance of Feature Engineering for Accelerated Design, offered a deep analysis of the role of data-driven methods in materials design. 

She shared how AI could help researchers better understand microstructure features—key components in determining material performance—and how this knowledge could be used to accelerate the design of new materials. Her remarks set the tone for the event, underscoring the growing importance of AI in driving innovation across materials science research.

In addition to fostering academic collaboration, the symposium also provided trainees with valuable career insights from industry and national lab leaders. A panel discussion on Career Perspectives Across Industry, Government, and Academic Sectors featured professionals from organizations like Argonne National Laboratory, Biomason, and one of Duke’s own healthcare ventures, restor3D. 

The panelists shared their experiences navigating careers in materials science, offering advice on the growing intersection between AI and materials research. Among the topics covered were emerging trends like the increasing demand for AI expertise in industries ranging from energy and manufacturing to healthcare, and speakers provided valuable advice on how trainees could position themselves for success in an increasingly interdisciplinary job market.

Looking ahead, the enthusiasm sparked by the symposium is already translating into future research collaborations. Cross-institutional interdisciplinary student teams competed to design research proposals connecting their disciplinary expertise and computational skills. The winning proposal team, Po-An Lin (Duke), Yasha Saxena (Duke), Jay Shah (University of Delaware), Tracy Asamoah (University of Chicago) will  further develop their project at upcoming hackathons and workshops. Faculty members from the participating institutions are also considering large-scale collaborations for future NSF funding opportunities.

“My favorite event was the Proposal Event,” said Wen Zhuang, a molecular engineering PhD candidate at the University of Chicago. “The brainstorming session with people from different backgrounds really broadened my perspective. It was a great opportunity to learn from each other and propose a new idea within just 30 minutes. It was challenging, but the joy of collaboration and the chance to open up new horizons made it a standout experience for me.”

The symposium was made possible by the support of Pratt Engineering, the Duke Materials Initiative, MEMS, and the NSF’s NRT-HDR program, which funds innovative graduate training programs designed to equip the next generation of scientists and engineers with the skills needed to drive advances in both fundamental research and practical applications. 

The event demonstrated the tremendous potential of combining AI with materials science, and the energy and excitement generated during the meeting suggests that the future of this interdisciplinary field is brighter than ever. As researchers continue to explore the use of machine learning in materials science, the symposium provided a powerful reminder of the importance of collaboration.