Adventures in Practical Population Inference - IPAM at UCLA

Explore the challenges and advancements in population inference for gravitational-wave astronomy in this 43-minute conference talk by Colm Talbot from the Massachusetts Institute of Technology. Dive into the computational complexities of extracting astrophysical and cosmological information from gravitational-wave observations, focusing on compact binary populations. Examine commonly used methods for astrophysical inference and their limitations. Discover novel approaches to mitigate these issues, including analytical integration, Monte Carlo integration, and density estimation techniques. Learn about selection effects, the observation-biased likelihood, and methods for evaluating selection functions. Gain insights into uncertainty quantification, scaled Gaussian mixture models, and continuous representations using Gaussian process regression and neural networks. Understand the challenges of comparing observations with predictions and explore potential solutions for more efficient analysis of growing gravitational-wave catalogs.