Learning effective physical laws for generating cosmological hydrodynamics with Lagrangian Deep Learning

TL;DR

This paper introduces Lagrangian Deep Learning (LDL), a physics-informed generative model that efficiently simulates cosmological hydrodynamics by learning effective physical laws, significantly reducing computational costs while maintaining high accuracy.

Contribution

The paper presents LDL, a novel physics-based deep learning approach that models cosmological outputs with few parameters, enabling fast and scalable simulations of complex hydrodynamical processes.

Findings

LDL reduces simulation costs by nearly four orders of magnitude.

LDL outperforms full hydrodynamical simulations at the same resolution.

The model uses only about 10 layers to connect initial conditions to outputs.

Abstract

The goal of generative models is to learn the intricate relations between the data to create new simulated data, but current approaches fail in very high dimensions. When the true data generating process is based on physical processes these impose symmetries and constraints, and the generative model can be created by learning an effective description of the underlying physics, which enables scaling of the generative model to very high dimensions. In this work we propose Lagrangian Deep Learning (LDL) for this purpose, applying it to learn outputs of cosmological hydrodynamical simulations. The model uses layers of Lagrangian displacements of particles describing the observables to learn the effective physical laws. The displacements are modeled as the gradient of an effective potential, which explicitly satisfies the translational and rotational invariance. The total number of learned parameters is only of order 10, and they can be viewed as effective theory parameters. We combine N-body solver FastPM with LDL and apply them to a wide range of cosmological outputs, from the dark matter to the stellar maps, gas density and temperature. The computational cost of LDL is nearly four orders of magnitude lower than the full hydrodynamical simulations, yet it outperforms it at the same resolution. We achieve this with only of order 10 layers from the initial conditions to the final output, in contrast to typical cosmological simulations with thousands of time steps. This opens up the possibility of analyzing cosmological observations entirely within this framework, without the need for large dark-matter simulations.