Seminar by Prof. David W. Hogg: " Is Machine Learning good or bad for Science?"

"IS Machine Learning good or bad for Science?"

Lanthieri Mansion, Vipava at 17:00 PM

Abstract:

Machine learning (ML) methods are having a huge impact across all of the sciences. However, ML has a strong ontology - in which only the data exist - and a strong epistemology - in which a model is considered good if it performs well on held-out training data. These philosophies are in strong conflict with both standard practices and key philosophies in the natural sciences. I identify some locations for ML in the natural sciences at which the ontology and epistemology are valuable. I also show that there are contexts in which the introduction of ML introduces strong, unwanted statistical biases. My partial answers I provide (to the question in my title) come from the particular perspective of physics.

Biography :

David W. Hogg is a Professor of Physics and Data Science in the Center for Cosmology and Particle Physics in the Department of Physics at New York University. He is also a Senior Research Scientist in the Astronomical Data Group in the Center for Computational Astrophysics of the Flatiron Institute in New York, and he spends a part of each year at the Max Planck Institute for Astronomy in Heidelberg, Germany, where he is a visiting researcher. His research has ranged across fundamental cosmological measurements, stellar dynamics in the Milky Way, precise measurement of stellar element abundances, and extra-solar planet discovery. His work includes a significant engineering component, in areas of instrument calibration, automated data analysis, and statistical inference. He is a developer and supporter of open-source software projects and open-data initiatives. Hogg earned his Ph.D. in physics from Caltech and his bachelor’s in physics from MIT. He was a long-term member of the Institute for Advanced Study in Princeton, N.J., where he became involved in the Sloan Digital Sky Survey.