Understanding the underlying processes that create data streams is a ubiquitous problem across the sciences. Whether it be mass-spectrometer or gravitational-wave data, developing and fitting models to data is key to understanding the underlying chemical, biophysical, or astrophysical processes.
Bayesian inference—the language of the gods and scientists—is the mathematical tool that enables physical inference from complicated data streams.
I am the PI on a project that has built a user-friendly Bayesian inference library—Bilby. With core developers Greg Ashton, Moritz Hübner, and Colm Talbot, we have built an easy-to-use software for performing Bayesian inference.
The primary purpose of the software was originally for performing astrophysical inference for gravitational-wave astronomy; work that is published in the Astrophysical Journal Supplements.
The software we developed is much more versatile than just being useful for gravitational-wave astronomy. To date, the Bilby code has been used for a number of other projects, including:
- model comparisons for gamma-ray burst afterglows
- characterising fast radio bursts
- understaning glitches in neutron stars
- x-ray timing of accreting neutron stars
- periodicity in active galactic nuclei
We anticipate the above list to grow. If you're keen to understand how Bilby can improve your life (or just your data analysis), then please don't hesitate to get in touch. Alternatively, we have a Bilby Slack channel, which is probably the quicket way to get help with the software.