ROME: Reduced modeling with extreme data

Learning fast computational models from data to accelerate energy breakthroughs.

A Department of Energy project developing Science Foundations for Energy Earthshots: Learning reduced models under extreme data conditions for design and rapid decision-making in complex systems.

About ROME

Our Research

Learning fast reduced models of physics simulations from distributed and streamed data.
Actively collecting data with rare-event simulation to learn models that are predictive even about extreme events.
Establishing trust in machine learning predictions through rigorous uncertainty quantification and multi-fidelity methods.

Meet the Team

Team