Massimiliano (Max) Lupo Pasini, an R&D data scientist from the Department of Energy’s Oak Ridge National Laboratory, was awarded the National Energy Research Scientific Computing Center’s (NERSC) High Performance Computing Achievement Award for High Impact Scientific Achievement for his work in “Groundbreaking contributions to scientific machine learning, particularly through the development of HydraGNN.”
Since 2013, the NERSC High Performance Computing Achievement Awards recognizes extraordinary contributions from junior scientists who use NERSC systems and resources in their research.
“I am very honored to have been selected as recipient of this award,” said Lupo Pasini. “NERSC is a widely utilized facility by researchers all over the world, and I knew that this would make the selection of the recipient a highly competitive process.”
A native of Italy, after earning a doctorate from Emory University, Lupo Pasini started working at ORNL in 2018 as a postdoctoral research associate. The following year, he became a full-time staff member and continued exploring deep learning and artificial intelligence algorithms for computational sciences and engineering applications.
“My work focuses on developing AI models to accelerate the design of materials with desired structural and functional properties using information from atomistic scale,” Lupo Pasini said. “I am very excited to work in this field because there is still a lot to unveil about how the materials’ characterization at atomistic scale affects important functional properties. I believe that AI can accelerate our understanding by breaking through computational barriers that otherwise would be unsurmountable, even for cutting edge leadership class supercomputing facilities.”
In 2022, Lupo Pasini and his team started developing HydraGNN, a scalable multi-headed graph neural network architecture. This type of neural network is designed to extract underlying physics correlations from the data to produce fast and accurate predictions of graph properties. The Artificial Intelligence for Scientific Discovery thrust of the AI Initiative at ORNL funded this effort to produce fast and accurate predictions of material properties using atomic information.
It was this work on NERSC supercomputing facilities that caught the attention of the award committee.
“NERSC allowed me to advance my research in materials science on multiple fronts,” said Lupo Pasini. “On the one hand, I was able to generate large volumes of highly accurate first-principles data. On the other hand, I was able to develop scalable, robust AI capabilities trained on the generated first-principles data to advance materials science technologies under the DOE portfolio to help maintain U.S. competitiveness in material science.”
As part of earning this distinction, Lupo Pasini’s work will be highlighted in NERSC press release, and he’ll be invited to present his work in the NERSC seminar series.
UT-Battelle manages ORNL for the Department of Energy’s Office of Science, the single largest supporter of basic research in the physical sciences in the United States. The Office of Science is working to address some of the most pressing challenges of our time. For more information, please visit energy.gov/science. – Mark Alewine
This Oak Ridge National Laboratory news article "Lupo Pasini receives NERSC Early Career Achievement Award" was originally found on https://www.ornl.gov/news
