ACME - Eryn

Tuesday 25 Nov 2025, 10:00 → 17:00 Europe/Paris

Argyro Sasli (University of Minnesota), Nikolaos Karnesis (Aristotle University of Thessaloniki)

Eryn is an advanced Bayesian tool designed for addressing complex data analysis challenges. It effectively manages multimodal posterior surfaces and can simultaneously handle different model types. It also provides reversible-jump sampling capabilities, making it ideal for scenarios where the number of models is uncertain. This tutorial will cover the fundamentals of Bayesian data analysis and demonstrate how to apply Eryn to solve data analysis problems for current and future Gravitational Wave detectors.

1 Introduction to Eryn

In this session we will explore the capabilities of Eryn by going through some basic example applications. We will discuss potential uses, and the flexibility of the software to allow for large-scale data analysis pipelines.

Speaker: Nikolaos Karnesis (Aristotle University of Thessaloniki)

ACME - ERYN.pdf

eryn_tutorial_acme.ipynb

2 Eryn as tool for data analysis in Astronomy

We will demonstrate how to perform parameter estimation (PE) for a Galactic Binary (GB) source. We will simulate the Gravitational Wave (GW) signal using the GBGPU package, and the electromagnetic (EM) light curve using the ellc package. Finally, we show a simple example of performing PE with a joint EM–GW likelihood, combining information from both datasets.

You may get the code and data from the dedicated git repository.

Speaker: Argyro Sasli (University of Minnesota)

data_gbgpu.npz

parameter_estimation_eryn_ucb_gw.ipynb