As a data deluge swamped traditional approaches, Astronomy, by necessity began to leverage AI/ML methods to facilitate discovery and inference at scale. Despite ongoing successes in the midstream (just after data is collected), I argue that our field sits just at the beginning of the real revolution where AI/ML becomes integrated and integral to the entire scientific workflow. From telescope and instrument design, to real-time control, to survey optimization, there is tremendous potential for AI/ML to improve the upstream efficiencies of data collection. Likewise, downstream from data are exciting opportunities for AI-enabled hypothesis generation and human-centered exploration. I will ground this talk with specific examples of ongoing work in simulation based inference, self-supervised ML, foundation models, neural compression, and reinforcement learning.
Biography
Joshua Bloom is a Miller Professor at the University of California, Berkeley. He teaches radiative processes, high-energy astrophysics, and a graduate-level "Python for Data Science" course. He has published over 300 refereed articles on time-domain transients events, machine learning, and telescope/insight automation. Bloom co-founded the Berkeley Institute for Data Science. He has been awarded the Data-Driven Discovery prize from the Gordon and Betty Moore Foundation and the Pierce Prize from the American Astronomical Society; he is also a former Sloan Fellow, Junior Fellow at the Harvard Society, and Hertz Foundation Fellow. He holds a PhD from Caltech and degrees from Harvard College and Cambridge University. He was co-founder and CTO of Wise.io, an AI application startup, acquired by GE Digital in 2016. His book on gamma-ray bursts, a technical introduction for physical scientists, was published by Princeton University Press.