# Super-resolving Dark Matter Halos using Generative Deep Learning

arXiv: https://arxiv.org/abs/2111.06393 · HTML: https://mlsanity.com/paper/2111.06393/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 38 references that are in the map.
- First posted on arXiv 11 Nov 2021.

## Builds on (most cited first, 11 of 11)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Conditional Generative Adversarial Nets](https://mlsanity.com/paper/1411.1784/) (2014)
- [Which Training Methods for GANs do actually Converge?](https://mlsanity.com/paper/1801.04406/) (2018)
- [Learning to Predict the Cosmological Structure Formation](https://mlsanity.com/paper/1811.06533/) (2018)
- [AI-assisted super-resolution cosmological simulations](https://mlsanity.com/paper/2010.06608/) (2020)
- [Fast and Accurate Non-Linear Predictions of Universes with Deep Learning](https://mlsanity.com/paper/2012.00240/) (2020)
- [AI-assisted super-resolution cosmological simulations II: Halo substructures, velocities and higher order statistics](https://mlsanity.com/paper/2105.01016/) (2021)
- [Learning neutrino effects in Cosmology with Convolutional Neural Networks](https://mlsanity.com/paper/1910.04255/) (2019)
- [From Dark Matter to Galaxies with Convolutional Neural Networks](https://mlsanity.com/paper/1910.07813/) (2019)
- [Learning the Evolution of the Universe in N-body Simulations](https://mlsanity.com/paper/2012.05472/) (2020)

## Abstract (arXiv)

Generative deep learning methods built upon Convolutional Neural Networks (CNNs) provide a great tool for predicting non-linear structure in cosmology. In this work we predict high resolution dark matter halos from large scale, low resolution dark matter only simulations. This is achieved by mapping lower resolution to higher resolution density fields of simulations sharing the same cosmology, initial conditions and box-sizes. To resolve structure down to a factor of 8 increase in mass resolution, we use a variation of U-Net with a conditional GAN, generating output that visually and statistically matches the high resolution target extremely well. This suggests that our method can be used to create high resolution density output over Gpc/h box-sizes from low resolution simulations with negligible computational effort.