Cobaya: code for Bayesian analysis of hierarchical physical models

By Jesús Torrado, A. Lewis. Journal of Cosmology and Astroparticle Physics, 2020

687 citations recorded by the source corpus.

We present , a general-purpose Bayesian analysis code aimed at models with complex internal interdependencies. Without the need for specific code by the user, interdependencies between different stages of a model pipeline are exploited for sampling efficiency: intermediate results are automatically cached, and parameters are grouped in blocks according to their dependencies and optimally sorted, taking into account their individual computational costs, so as to minimize the cost of their variation during sampling, thanks to a novel algorithm. Cobaya allows exploration of posteriors using a range of Monte Carlo samplers, and also has functions for maximization and importance-reweighting of Monte Carlo samples with new priors and likelihoods. Cobaya is written in Python in a modular way that allows for extendability, use of calculations provided by external packages, and dynamical reparameterization without modifying its source. It can exploit hybrid OpenMP/MPI parallelization, and has sub-millisecond overhead per posterior evaluation. Though Cobaya is a general purpose statistical framework, it includes interfaces to a set of cosmological Boltzmann codes and likelihoods (the latter being agnostic with respect to the choice of the former), and automatic installers for external dependencies.

Subfields: Computational Mechanics, Statistical and Nonlinear Physics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/cd2a92d6ae1c42fe80f936cfc0653a333f51c727

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

1. J. Carron
2. Z. Vlah
3. M. Mirmelstein
4. M. White
5. Shi-Fan Chen
6. A. Lewis
7. Adam D. Myers
8. C. Rockosi

In Statistical and Nonlinear Physics

1. Z. Vlah
2. Shi-Fan Chen
3. M. White
4. Jesús Torrado
5. A. Banday
6. K. G'orski
7. J. Carron
8. M. Mirmelstein

In Statistics and Probability

1. J. Carron
2. M. White
3. Z. Vlah
4. M. Mirmelstein
5. Shi-Fan Chen
6. A. Lewis
7. Sergey E. Koposov
8. Risa Wechsler

Citations in the corpus

• Improved reconstruction of a stochastic gravitational wave background with LISA (2020, methodology): “ in which the foreground amplitude is t simultaneously for all bins, as the noise parameters are. In the examples below we have used PolyChord [58,59] as the MC sampler, via its interface with Cobaya [55]. The results were analyzed and the contour plots were produced using GetDist [60]. In g.11we sh”
• Cosmological parameters derived from the final Planck data release (PR4) (2023, methodology): “To quantify the agreement between the data and the model, we computed the χ 2 values with respect to the best-fit model for each of the data sets using Cobaya (Torrado & Lewis 2021) with its adaptive, speed-hierarchy-aware MCMC sampler (Lewis & Bridle 2002; Lewis 2013).”
• A new analysis of galaxy 2-point functions in the BOSS survey, including full-shape information and post-reconstruction BAO (2021, methodology): “We acknowledge the use of Cobaya [102, 103], Class [50], GetDist [104] and velocileptors [17] and thank their authors for making these products public.”
• The Atacama Cosmology Telescope: A Measurement of the DR6 CMB Lensing Power Spectrum and Its Implications for Structure Growth (2023, methodology): “We also acknowledge use of the matplotlib Hunter (2007) package and the Python Image Library for producing plots in this paper, and use of the Boltzmann code CAMB Lewis et al. (2000) for calculating theory spectra and use of the GetDist (Lewis 2019), Cobaya (Torrado & Lewis 2021) for sampling and li”
• DESI 2024: reconstructing dark energy using crossing statistics with DESI DR1 BAO data (2024, methodology): “EE ) [88], as implemented in cobaya [89].”
• The Atacama Cosmology Telescope: DR6 Gravitational Lensing Map and Cosmological Parameters (2023, methodology): “We also acknowledge use of the matplotlib (Hunter 2007) package and the Python Image Library for producing plots in this paper, use of 18 http://www.astropy.org the Boltzmann code CAMB (Lewis et al. 2000) for calculating theory spectra, and use of the GetDist (Lewis 2019), Cobaya (Torrado & Lewis 20”
• CMB lensing from Planck PR4 maps (2022, methodology): “We use the Markov Chain Monte Carlo sampling method with Cobaya7 [38] to estimate cosmological parameters from the calculated likelihoods, and analyse the resulting chains using GetDist8 [39].”
• Validation of the Scientific Program for the Dark Energy Spectroscopic Instrument (2023, methodology): “We use the publicly available code C obaya (Torrado & Lewis 2021) to fit the parameters to the data and infer their posterior distribution.”
• On the Evolution of the Hubble Constant with the SNe Ia Pantheon Sample and Baryon Acoustic Oscillations: A Feasibility Study for GRB-Cosmology in 2030 (2022, methodology): “More specifically, after the minimization of χ2, we extract the H0 value in each redshift bin, via the Cobaya code [260].”