Learning cosmology from nearest neighbour statistics

Abstract

Extracting cosmological parameters from galaxy and halo catalogues to sub-per cent level accuracy is an important aspect of modern cosmology, especially in view of ongoing and upcoming surveys such as Euclid, DESI, and LSST. While traditional two-point statistics have been known to be suboptimal for this task, recently proposed k-nearest neighbour (kNN) based summary statistics have demonstrated a tighter constraining power. Building on the kNN statistics, we introduced a new field-level representation of discrete halo catalogues: NN distance maps. We employed this technique on the halo catalogues obtained from Quijote N-body simulation suites. By combining these maps with kNN-based summary statistics, we trained a hybrid neural network to infer cosmological parameters, showing that the resulting constraints achieve state-of-the-art accuracy, comparable to the best existing methods. In addition, our hybrid framework is 5 − 10 times more computationally efficient than some of the existing point-cloud-based ML methods.