Deep Learning for Cosmological Parameter Inference from Dark Matter Halo Density Field

We propose a lightweight deep convolutional neural network (lCNN) to estimate cosmological parameters from simulated three-dimensional DM halo distributions and associated statistics. The training dataset comprises 2000 realizations of a cubic box with a side length of 1000 h^-1 Mpc, and interpolated over a cubic grid of 300³ voxels, with each simulation produced using 512³ DM particles and 512³ neutrinos. Under the flat CDM model, simulations vary standard six cosmological parameters including ₘ, b, h, nₛ, ₈, w, along with the neutrino mass sum, M_. We find that: 1) within the framework of lCNN, extracting large-scale structure information is more efficient from the halo density field compared to relying on the statistical quantities including the power spectrum, the two-point correlation function, and the coefficients from wavelet scattering transform; 2) combining the halo density field with its Fourier transformed counterpart enhances predictions, while augmenting the training dataset with measured statistics further improves performance; 3) achieving high accuracy in inferring ₘ, h, nₛ, and ₈ by the neural network model, while being inefficient in predicting b, M_ and w; 4) compared to the simple random forest network trained with three statistical quantities, lCNN yields unbiased estimations with reduced statistical errors: approximately 33. 3\% for ₘ, 20. 0\% for h, 8. 3\% for nₛ, and 40. 0\% for ₈. Our study emphasizes this lCNN-based novel approach in extracting large-scale structure information and estimating cosmological parameters.