A Hybrid Deep Learning Approach to Cosmological Constraints from Galaxy Redshift Surveys
Abstract
We present a deep machine learning (ML)-based technique for accurately determining σ8 and Ωm from mock 3D galaxy surveys. The mock surveys are built from the AbacusCosmos suite of N-body simulations, which comprises 40 cosmological volume simulations spanning a range of cosmological parameter values, and we account for uncertainties in galaxy formation scenarios through the use of generalized halo occupation distributions (HODs). We explore a trio of ML models: a 3D convolutional neural network (CNN), a power spectrum-based fully connected network, and a hybrid approach that merges the two to combine physically motivated summary statistics with flexible CNNs. We describe best practices for training a deep model on a suite of matched-phase simulations, and we test our model on a completely independent sample that uses previously unseen initial conditions, cosmological parameters, and HOD parameters. Despite the fact that the mock observations are quite small (∼0.07 h-3 Gpc3) and the training data span a large parameter space (six cosmological and six HOD parameters), the CNN and hybrid CNN can constrain estimates of σ8 and Ωm to ∼3% and ∼4%, respectively.
- Publication:
-
The Astrophysical Journal
- Pub Date:
- February 2020
- DOI:
- 10.3847/1538-4357/ab5f5e
- arXiv:
- arXiv:1909.10527
- Bibcode:
- 2020ApJ...889..151N
- Keywords:
-
- Cosmological parameters from large-scale structure;
- Large-scale structure of the universe;
- Astrostatistics;
- Cosmology;
- 340;
- 902;
- 1882;
- 343;
- Astrophysics - Cosmology and Nongalactic Astrophysics
- E-Print:
- Submitted to The Astrophysical Journal