Correlation between several cosmological quantities using machine learning model based on cosmological simulation data

Yuxuan Cao, Kaile Wang, Saiyang Zhang, Joohyun Lee

Goals

Study the correlation between SMBH merger history and DM halo merger history, i.e. rebuild SMBH growth history from host halo merger history + their properties

Models to be used:

regression(e.g. multi-variate), classification (e.g. random forest), machine learning

Variables in the problem:

Dependent :

BH Merger rate(host halo properties, BH properties)
BH Merger mass ratio(host halo properties, BH properties)

Independent:

Host halo properties: host merger rate, host merger mass ratio, mass, spin, central density, maximum rotation velocity, …
BH properties: chirp mass, spin angle, mass, velocity, …

Links to data

IllustrisTNG100-1 cosmological simulation

Data description

One of the state-of-the-art cosmological hydrodynamic simulations of galaxy formation
Helps to understand when and how galaxies evolve into the structures that are observed

References

https://ui.adsabs.harvard.edu/abs/2023MNRAS.518.2123Z/abstract
https://ui.adsabs.harvard.edu/abs/2023MNRAS.523L..69Z/abstract

Sylvanzsy / SDS384_Spr24