Workflow Provenance in the Lifecycle of Scientific Machine Learning
This work addresses reproducibility, explainability, and data understanding for scientists in domains like geoscience and health science, though it is incremental as it applies existing provenance techniques to scientific ML.
The authors tackled the challenge of supporting comprehensive data analyses in scientific machine learning by leveraging workflow provenance techniques, resulting in a system that enables integrated queries with low overhead (<1%), high scalability, and up to an order of magnitude query acceleration in evaluations.
Machine Learning (ML) has already fundamentally changed several businesses. More recently, it has also been profoundly impacting the computational science and engineering domains, like geoscience, climate science, and health science. In these domains, users need to perform comprehensive data analyses combining scientific data and ML models to provide for critical requirements, such as reproducibility, model explainability, and experiment data understanding. However, scientific ML is multidisciplinary, heterogeneous, and affected by the physical constraints of the domain, making such analyses even more challenging. In this work, we leverage workflow provenance techniques to build a holistic view to support the lifecycle of scientific ML. We contribute with (i) characterization of the lifecycle and taxonomy for data analyses; (ii) design principles to build this view, with a W3C PROV compliant data representation and a reference system architecture; and (iii) lessons learned after an evaluation in an Oil & Gas case using an HPC cluster with 393 nodes and 946 GPUs. The experiments show that the principles enable queries that integrate domain semantics with ML models while keeping low overhead (<1%), high scalability, and an order of magnitude of query acceleration under certain workloads against without our representation.