Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 400 - Multiple Aspects of Bayesian Model Selection and Variable Selection in Linear and Nonlinear Models
Type: Topic Contributed
Date/Time: Wednesday, August 5, 2020 : 1:00 PM to 2:50 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #313912
Title: Probabilistic Machine Learning for Uncertainty Quantification of Neutron Star Radius
Author(s): Debdeep Pati* and Anirban Battacharya and Yeunhwan Lim and Jeremy Holt
Companies: Texas A&M University and Texas A&M University and Max Planck Institut für Kernphysik, Technische Universität Darmstadt and Texas A&M University
Keywords: Nuclear Physics; uncertainty quantification; neutron star
Abstract:

With the advancement of probing technology, there has been rapid progress in accurately quantifying the radius and mass distribution of neutron stars. Although the mass distribution of neutron stars has been calibrated with a significantly high precision, precisely estimating the radius continues to pose a stiff challenge. High quality datasets from X-ray satellites have propelled significant progress in theoretical modeling, placing the radius in the 9.9 ? 11.2 km range and shrinking the uncertainty ranges due to a better understanding of the sources of systematic errors. The guiding equation that relates the observables (scattering angle probabilities) to the mass and the radius is called the equation of state, and recovering the mass and radius from noisy observations around the equation of state defines a statistical inverse problem. In this work, we propose a novel probabilistic mechanism to solve this inverse problem that takes into account various sources of uncertainty, integrates them using the laws of Physics, and provides more accurate uncertainty quantification for the neutron star radius.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program