As part of its commitment to advancing science through the use of artificial intelligence, the U.S. Department of Energy recently announced more than $320 million in investments toward projects and awards to accelerate AI capabilities.

These investments will support the Genesis Mission, a broader initiative by the U.S. Department of Energy to develop an integrated platform connecting the world’s best supercomputers, experimental facilities, AI systems, and unique datasets across every major scientific domain.

Yulia R. Gel, a professor in the Department of Statistics, is a key member of one of these new projects: the LEarning-Accelerated Domain Science (LEADS) Institute, which aims to make scientific machine learning accessible to domain scientists.

Research in the LEADS Institute will contribute to the development of novel algorithms for accurate and efficient exploration of large-scale, complex data and real-time information extraction using digital-twin-assisted optimal control.

Gel’s own research agenda focuses on statistical topological and geometric algorithms, specifically working on efficient graph learning, graph-based AI, and the associated uncertainty quantification.

“I am excited to contribute to the LEADS initiative, as it breaks down the traditional disciplinary boundaries and brings the statistical and mathematical foundations to the forefront of innovations in scientific computing,” said Gel, who joined the Virginia Tech faculty in 2024 after a stint as program officer for the National Science Foundation. “I also view it as an opportunity to redefine and highlight the unique role modern statistical science plays not only in scientific machine learning, but the AI breakthroughs in general.”

LEADS, which features researchers from 14 different institutions across academia and the national laboratory complex, is now one of three Scientific Discovery Through Advanced Computing (SciDAC) institutes supported by the Department of Energy’s Advanced Scientific Computing Research program. The two existing SciDAC institutes include:

  • FASTMath: Frameworks, Algorithms, and Scalable Technologies for Mathematics
  • RAPIDS: SciDac Institute for Computer Science and Data

Panos Stinis, the leader of the Computational Mathematics group at Pacific Northwest National Laboratory, will direct the LEADS Institute. According to Stinis, “LEADS will bridge the gap between scientific machine learning experts and domain scientists, enabling the development of state-of-the-art, highly customized, accurate, and efficient algorithms that leverage the vast domain knowledge within the Department of Energy complex.”

Contact:

Tags

Share this story