CRA -Machine Learning + Astro Seminar : Alec Gunny (MIT)

Title: Accelerating ML workflows in gravitational wave astronomy

Abstract: Commercial applications of machine learning (ML) have grown substantially in recent years, driven in no small part by an explosion in the ecosystem of so-called MLOps tools that have simplified the development and deployment of common ML use cases using domain-specific best practices. As the canon of ML applications in gravitational wave astronomy proliferates, it is worthwhile to consider what a similar ecosystem, optimized for this unique scientific and computing environment, might look like, and how it might help address the highest-value science goals in the coming years. In this talk, we will discuss some lessons learned from developing a few specific ML applications in this context, and introduce a growing suite of tools designed to encode these lessons and accelerate the pace at which robust, scalable ML applications can be built, tested and put into production.