Creating a more sustainable future will increasingly rely on the tools of artificial intelligence (AI) and machine learning.
But training leaders who can guide us into this future — by using AI to discover and design new materials for more efficient batteries, for example — requires a new educational approach.
For the past several years, AIMEMS has trained more than 16 graduate students on both the tools of AI and machine learning and the importance of interdisciplinarity collaboration, communication, and outreach.
The program’s unique approach — giving each graduate student a mentor from UChicago, Argonne National Laboratory, and industry, as well as developing new courses and outreach programs — has been so successful that those involved think it could be a blueprint for integrating AI across disciplines. (Read about three student experiences here.)
Together with similar programs from Duke University and the University of Illinois at Urbana-Champaign, AIMEMS faculty and staff wrote a Focus paper for the journal Science Advances, detailing their approach and successes.
“Adoption of AI and machine learning in scientific domains has been slow, and the limiting factor is the workforce,” said Prof. Juan de Pablo, executive vice president for science, innovation, national laboratories and global initiatives, and PI of the grant that funds the program. “We see this as an incredible opportunity to teach graduate students how to break down department siloes and collaborate across disciplines to accelerate the discovery and design of materials for sustainability. It’s an approach that can be applicable to many fields.”
Breaking down silos
The program focuses on the field of materials informatics, which integrates AI/machine learning and computational methodologies with materials science for rapid materials discovery, understanding, and design. Materials informatics has the potential to affect a wide range of fields, including batteries, water, catalysts, and nanomedicine.
But reaching that potential will require scientists and engineers to have both the technical skills of data analytics and AI/machine learning along with the domain-specific knowledge of materials science. They must also learn to work across disciplines, communicate their research, and provide outreach to the next generation of potential scientists.
Since 2021, AIMEMS, funded by the National Science Foundation’s Research Traineeship program, has recruited three cohorts of the program to focus their education and efforts toward the wide-ranging issue of sustainability. Graduate students from across molecular engineering, computer science, physical science, and social science have undergone boot camps to learn new technical skills and taken new courses on advanced materials characterization and AI/machine learning.
Along the way, each student has had three mentors: a UChicago faculty member, a scientist from Argonne, and an industry professional. These mentors guide graduate students on research, professional development, teamwork, and entrepreneurship.
AIMEMS trainees have also developed outreach events for K-12 students from schools that have predominantly low-income underrepresented minority populations. Events have included lab tours, hands-on activities, and lessons on how the idea of sustainability might affect their local communities.
“Our students have been really engaged and excited to make their science interactive for high school and middle school students,” said Jennifer Nolan, project coordinator for AIMEMS. “It helps develop their communication skills and it helps these young students learn about STEM. Each experience has been a real highlight of the program.”
A new model to integrate AI into science
In the paper, AIMEMS faculty and staff offer key steps to making such programs work: providing core training in AI/machine learning, breaking down department silos to encourage students to work in collaborative teams, and prioritizing partnerships with national labs and industry.
“We found that we need to first provide coursework or boot camps to get everyone up to speed on AI and machine learning,” said Prof. Junhong Chen, a leader of the AIMEMS program who is also a lead water strategist at Argonne National Laboratory. “And then we need to facilitate interdisciplinary projects, but that is how students learn the most. Solutions across departments will always be better than solutions from one discipline.”
The program has been so successful that students and faculty have been invited to share their experiences at National Science Foundation events. “Students talk about the program with a sense of ownership and pride,” Nolan said. “It has created a community that they have fueled with their own creativity and passion.”
Ultimately, faculty would like to expand this model to integrate AI into other disciplines across the university. “We hope those in our community will be inspired to create a similar educational approach to train students to tackle more global problems,” Chen said.