skip to main content

This content will become publicly available on April 1, 2023

Title: The CAMELS Multifield Data Set: Learning the Universe’s Fundamental Parameters with Artificial Intelligence
Abstract We present the Cosmology and Astrophysics with Machine Learning Simulations (CAMELS) Multifield Data set (CMD), a collection of hundreds of thousands of 2D maps and 3D grids containing many different properties of cosmic gas, dark matter, and stars from more than 2000 distinct simulated universes at several cosmic times. The 2D maps and 3D grids represent cosmic regions that span ∼100 million light-years and have been generated from thousands of state-of-the-art hydrodynamic and gravity-only N -body simulations from the CAMELS project. Designed to train machine-learning models, CMD is the largest data set of its kind containing more than 70 TB of data. In this paper we describe CMD in detail and outline a few of its applications. We focus our attention on one such task, parameter inference, formulating the problems we face as a challenge to the community. We release all data and provide further technical details at https://camels-multifield-dataset.readthedocs.io .
Authors:
; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; more » ; ; ; ; ; ; ; ; « less
Award ID(s):
2108944
Publication Date:
NSF-PAR ID:
10331321
Journal Name:
The Astrophysical Journal Supplement Series
Volume:
259
Issue:
2
Page Range or eLocation-ID:
61
ISSN:
0067-0049
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    We presentGIGANTES, the most extensive and realistic void catalog suite ever released—containing over 1 billion cosmic voids covering a volume larger than the observable universe, more than 20 TB of data, and created by running the void finderVIDEonQUIJOTE’s halo simulations. TheGIGANTESsuite, spanning thousands of cosmological models, opens up the study of voids, answering compelling questions: Do voids carry unique cosmological information? How is this information correlated with galaxy information? Leveraging the large number of voids in theGIGANTESsuite, our Fisher constraints demonstrate voids contain additional information, critically tightening constraints on cosmological parameters. We use traditional void summary statistics (void size function, void density profile) and the void autocorrelation function, which independently yields an error of 0.13 eV on ∑mνfor a 1h−3Gpc3simulation, without cosmic microwave background priors. Combining halos and voids we forecast an error of 0.09 eV from the same volume, representing a gain of 60% compared to halos alone. Extrapolating to next generation multi-Gpc3surveys such as the Dark Energy Spectroscopic Instrument, Euclid, the Spectro-Photometer for the History of the Universe and Ices Explorer, and the Roman Space Telescope, we expect voids should yield an independent determination of neutrino mass. Crucially,GIGANTESis the first void catalog suite expressly built for intensivemore »machine-learning exploration. We illustrate this by training a neural network to perform likelihood-free inference on the void size function, giving a ∼20% constraint on Ωm. Cosmology problems provide an impetus to develop novel deep-learning techniques. WithGIGANTES, machine learning gains an impressive data set, offering unique problems that will stimulate new techniques.

    « less
  2. Abstract

    Data-driven design of mechanical metamaterials is an increasingly popular method to combat costly physical simulations and immense, often intractable, geometrical design spaces. Using a precomputed dataset of unit cells, a multiscale structure can be quickly filled via combinatorial search algorithms, and machine learning models can be trained to accelerate the process. However, the dependence on data induces a unique challenge: An imbalanced dataset containing more of certain shapes or physical properties than others can be detrimental to the efficacy of the approaches and any models built on those sets. In answer, we posit that a smaller yet diverse set of unit cells leads to scalable search and unbiased learning. To select such subsets, we propose METASET, a methodology that 1) uses similarity metrics and positive semi-definite kernels to jointly measure the closeness of unit cells in both shape and property space, and 2) incorporates Determinantal Point Processes for efficient subset selection. Moreover, METASET allows the trade-off between shape and property diversity so that subsets can be tuned for various applications. Through the design of 2D metamaterials with target displacement profiles, we demonstrate that smaller, diverse subsets can indeed improve the search process as well as structural performance. We alsomore »apply METASET to eliminate inherent overlaps in a dataset of 3D unit cells created with symmetry rules, distilling it down to the most unique families. Our diverse subsets are provided publicly for use by any designer.

    « less
  3. Abstract

    From 1000 hydrodynamic simulations of the CAMELS project, each with a different value of the cosmological and astrophysical parameters, we generate 15,000 gas temperature maps. We use a state-of-the-art deep convolutional neural network to recover missing data from those maps. We mimic the missing data by applying regular and irregular binary masks that cover either 15% or 30% of the area. We quantify the reliability of our results using two summary statistics: (1) the distance between the probability density functions, estimated using the Kolmogorov–Smirnov (K-S) test, and (2) the 2D power spectrum. We find an excellent agreement between the model prediction and the unmasked maps when using the power spectrum: better than 1% fork< 20hMpc−1for any irregular mask. For regular masks, we observe a systematic offset of ∼5% when covering 15% of the maps, while the results become unreliable when 30% of the data is missing. The observed K-S testp-values favor the null hypothesis that the reconstructed and the ground-truth maps are drawn from the same underlying distribution when irregular masks are used. For regular-shaped masks, on the other hand, we find a strong evidence that the two distributions do not match each other. Finally, we use the model,more »trained on gas temperature maps, to inpaint maps from fields not used during model training. We find that, visually, our model is able to reconstruct the missing pixels from the maps of those fields with great accuracy, although its performance using summary statistics depends strongly on the considered field.

    « less
  4. Abstract Background Cryo-EM data generated by electron tomography (ET) contains images for individual protein particles in different orientations and tilted angles. Individual cryo-EM particles can be aligned to reconstruct a 3D density map of a protein structure. However, low contrast and high noise in particle images make it challenging to build 3D density maps at intermediate to high resolution (1–3 Å). To overcome this problem, we propose a fully automated cryo-EM 3D density map reconstruction approach based on deep learning particle picking. Results A perfect 2D particle mask is fully automatically generated for every single particle. Then, it uses a computer vision image alignment algorithm (image registration) to fully automatically align the particle masks. It calculates the difference of the particle image orientation angles to align the original particle image. Finally, it reconstructs a localized 3D density map between every two single-particle images that have the largest number of corresponding features. The localized 3D density maps are then averaged to reconstruct a final 3D density map. The constructed 3D density map results illustrate the potential to determine the structures of the molecules using a few samples of good particles. Also, using the localized particle samples (with no background) to generate themore »localized 3D density maps can improve the process of the resolution evaluation in experimental maps of cryo-EM. Tested on two widely used datasets, Auto3DCryoMap is able to reconstruct good 3D density maps using only a few thousand protein particle images, which is much smaller than hundreds of thousands of particles required by the existing methods. Conclusions We design a fully automated approach for cryo-EM 3D density maps reconstruction (Auto3DCryoMap). Instead of increasing the signal-to-noise ratio by using 2D class averaging, our approach uses 2D particle masks to produce locally aligned particle images. Auto3DCryoMap is able to accurately align structural particle shapes. Also, it is able to construct a decent 3D density map from only a few thousand aligned particle images while the existing tools require hundreds of thousands of particle images. Finally, by using the pre-processed particle images,Auto3DCryoMap reconstructs a better 3D density map than using the original particle images.« less
  5. Abstract Data-driven design of mechanical metamaterials is an increasingly popular method to combat costly physical simulations and immense, often intractable, geometrical design spaces. Using a precomputed dataset of unit cells, a multiscale structure can be quickly filled via combinatorial search algorithms, and machine learning models can be trained to accelerate the process. However, the dependence on data induces a unique challenge: an imbalanced dataset containing more of certain shapes or physical properties can be detrimental to the efficacy of data-driven approaches. In answer, we posit that a smaller yet diverse set of unit cells leads to scalable search and unbiased learning. To select such subsets, we propose METASET, a methodology that (1) uses similarity metrics and positive semi-definite kernels to jointly measure the closeness of unit cells in both shape and property spaces and (2) incorporates Determinantal Point Processes for efficient subset selection. Moreover, METASET allows the trade-off between shape and property diversity so that subsets can be tuned for various applications. Through the design of 2D metamaterials with target displacement profiles, we demonstrate that smaller, diverse subsets can indeed improve the search process as well as structural performance. By eliminating inherent overlaps in a dataset of 3D unit cellsmore »created with symmetry rules, we also illustrate that our flexible method can distill unique subsets regardless of the metric employed. Our diverse subsets are provided publicly for use by any designer.« less