Investigating Cosmological GAN Emulators Using Latent Space Interpolation

Abstract

Generative adversarial networks (GANs) have been recently applied as a novel emulation technique for large scale structure simulations. Recent results show that GANs can be used as a fast, efficient and computationally cheap emulator for producing novel weak lensing convergence maps as well as cosmic web data in 2-D and 3-D. However, like any algorithm, the GAN approach comes with a set of limitations, such as an unstable training procedure and the inherent randomness of the produced outputs. In this work we employ a number of techniques commonly used in the machine learning literature to address the mentioned limitations. In particular, we train a GAN to produce both weak lensing convergence maps and dark matter overdensity field data for multiple redshifts, cosmological parameters and modified gravity models. In addition, we train a GAN using the newest Illustris data to emulate dark matter, gas and internal energy distribution data simultaneously. Finally, we apply the technique of latent space interpolation to control which outputs the algorithm produces. Our results indicate a 1-20% difference between the power spectra of the GAN-produced and the training data samples depending on the dataset used and whether Gaussian smoothing was applied. Finally, recent research on generative models suggests that such algorithms can be treated as mappings from a lower-dimensional input (latent) space to a higher dimensional (data) manifold. We explore such a theoretical description as a tool for better understanding the latent space interpolation procedure.

arXiv comment20 pages, 23 figures, 3 tables

References in corpus (36)

• XGBoost: A Scalable Tree Boosting System
• Conditional Generative Adversarial Nets
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• The cosmological simulation code GADGET-2
• Properties of galaxies reproduced by a hydrodynamic simulation
• InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
• The Illustris Simulation: Public Data Release
• VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning
• Solving Large Scale Structure in Ten Easy Steps with COLA
• nbodykit: an open-source, massively parallel toolkit for large-scale structure
• Learning to Predict the Cosmological Structure Formation
• Euclid preparation: II. The EuclidEmulator -- A tool to compute the cosmology dependence of the nonlinear matter power spectrum
• Cosmology with the shear-peak statistics
• Large-scale Bias and Efficient Generation of Initial Conditions for Non-Local Primordial Non-Gaussianity
• Quantifying the Carbon Emissions of Machine Learning
• Non-Gaussian information from weak lensing data via deep learning
• CosmoGAN: creating high-fidelity weak lensing convergence maps using Generative Adversarial Networks
• Cosmic Emulation: Fast Predictions for the Galaxy Power Spectrum
• COLA with scale-dependent growth: applications to screened modified gravity models
• Fast cosmic web simulations with generative adversarial networks
• Cosmology with Minkowski functionals and moments of the weak lensing convergence field
• Super-resolution emulator of cosmological simulations using deep physical models
• Emulators for the non-linear matter power spectrum beyond $\Lambda$CDM
• Constraining neutrino mass with weak lensing Minkowski Functionals
• On the dissection of degenerate cosmologies with machine learning
• The Role of Machine Learning in the Next Decade of Cosmology
• Do dark matter halos explain lensing peaks?
• Generative deep fields: arbitrarily sized, random synthetic astronomical images through deep learning
• Angular momentum loss for eccentric compact binary in screened modified gravity
• GANs May Have No Nash Equilibria
• Encoding large scale cosmological structure with Generative Adversarial Networks
• Distinguishing general relativity and $f(R)$ gravity with the gravitational lensing Minkowski functionals
• The Riemannian Geometry of Deep Generative Models
• CMB-GAN: Fast Simulations of Cosmic Microwave background anisotropy maps using Deep Learning
• Dark Energy Survey Year 1 Results: Wide field mass maps via forward fitting in harmonic space
• Comparison of Generative Adversarial Networks Architectures Which Reduce Mode Collapse

Cited by in corpus (10)

• Large-scale dark matter simulations
• Super-resolution emulator of cosmological simulations using deep physical models
• The CAMELS Multifield Dataset: Learning the Universe's Fundamental Parameters with Artificial Intelligence
• From EMBER to FIRE: predicting high resolution baryon fields from dark matter simulations with Deep Learning
• MillimeterDL: Deep Learning Simulations of the Microwave Sky
• Bayesian deep learning for cosmic volumes with modified gravity
• Neural style transfer of weak lensing mass maps
• Stochastic Super-resolution of Cosmological Simulations with Denoising Diffusion Models
• Reconstructing the mid-infrared spectra of galaxies using ultraviolet to submillimeter photometry and Deep Generative Networks
• nuGAN: Generative Adversarial Emulator for Cosmic Web with Neutrinos