DescriptionMachine learning (ML) is rapidly changing the field of scientific computing. However, there exist key research gaps that limit the impact of current ML methods on scientific problems. These gaps were identified and discussed at a recent Department of Energy Basic Research Needs Workshop on Scientific Machine Learning. This workshop identified six priority research directions to increase the impact of ML on scientific problems. In this talk, I will review these six areas and present recent results from Pacific Northwest National Laboratory in domain-aware machine learning, one of the six priority research directions.