What can machine learning tell us about the background expansion of the Universe?

Arjona, Rubén; Nesseris, Savvas

doi:10.1103/physrevd.101.123525

Cited by 62 publications

(48 citation statements)

References 89 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[45][46][47][48][49][50][51][52]. Recently the non-parametric approach finds relevance in the reconstruction of kinematical quantities like the deceleration parameter q [53][54][55][56][57][58][59][60] and the jerk parameter j [61].…”

Section: Introductionmentioning

confidence: 99%

Nonparametric reconstruction of interaction in the cosmic dark sector

Mukherjee

Banerjee

2021

Phys. Rev. D

View full text Add to dashboard Cite

The possibility of a non-gravitational interaction between the dark matter and the dark energy has been reconstructed using some recent datasets. The crucial aspect is that the interaction is not parametrized at the outset, but rather reconstructed directly from the data in a non-parametric way. The Cosmic Chronometer Hubble data, the Pantheon Supernova compilation of CANDELS and CLASH Multy-Cycle Treasury programs obtained by the HST, and the Baryon Acoustic Oscillation Hubble data have been considered in this work. The widely accepted Gaussian Process is used for the reconstruction. The results clearly indicate that a no interaction scenario is quite a possibility. Also, the interaction, if any, is not really significant at the present epoch. The direction of the flow of energy is clearly from the dark energy to the dark matter which is consistent with the thermodynamic requirement.

show abstract

Section: Introductionmentioning

confidence: 99%

Nonparametric reconstruction of interaction in the cosmic dark sector

Mukherjee

Banerjee

2021

Phys. Rev. D

View full text Add to dashboard Cite

show abstract

“…) respectively, we can solve for Ω m,0 similarly to the Om H (z) test, see Arjona & Nesseris (2020b).…”

Section: The Distance Null Testsmentioning

confidence: 99%

Euclid: Forecast constraints on consistency tests of the $Λ$CDM model

Nesseris,

Sapone,

Martinelli

et al. 2021

Preprint

View full text Add to dashboard Cite

Context:The standard cosmological model is based on the simplifying assumptions of a spatially homogeneous and isotropic universe on large scales. An observational detection of their violation, at any redshift, would immediately indicate the breakdown of the aforementioned assumptions or presence of new physics. Aims: We quantify the ability of the Euclid mission, together with contemporary surveys, to improve the current sensitivity of null tests of the canonical cosmological constant Λ and cold dark matter (ΛCDM) model, in the redshift range 0 < z < 1.8. Methods: We consider both currently available data and simulated Euclid and external data products based on a ΛCDM fiducial model, an evolving dark energy model assuming the Chevallier-Polarski-Linder (CPL) parametrization or an inhomogeneous Lemaître-Tolman-Bondi model with a cosmological constant Λ (ΛLTB), and carry out two separate, albeit complementary, analyses: a machine learning reconstruction based on genetic algorithms and a theory-agnostic parametric approach based on polynomial reconstruction and binning of the data, in order not to make assumptions about any particular model. Results: We find that using the machine learning approach Euclid can (in combination with external probes) improve current constraints on null tests of the ΛCDM by approximately a factor of two to three, while in the case of the binning approach, Euclid can provide tighter constraints than the genetic algorithms by a further factor of two in the case of the ΛCDM mock, albeit in certain cases may be biased against or missing some features of models far from ΛCDM, as is the case with the CPL and ΛLTB mocks. Conclusions: Our analysis highlights the importance of synergies between Euclid and other surveys, which are crucial to provide tighter constraints over an extended redshift range, for a plethora of different consistency tests of some of the main assumptions of the current cosmological paradigm.

show abstract

“…As the amount of observational data increases, the choice of data analysis methods becomes more crucial. Therefore, computational methods, including machine learning, have been incorporated into the cosmological field in recent years [1][2][3].…”

Section: Introductionmentioning

confidence: 99%

Observational Cosmology with Artificial Neural Networks

2022

View full text Add to dashboard Cite

In cosmology, the analysis of observational evidence is very important when testing theoretical models of the Universe. Artificial neural networks are powerful and versatile computational tools for data modelling and have recently been considered in the analysis of cosmological data. The main goal of this paper is to provide an introduction to artificial neural networks and to describe some of their applications to cosmology. We present an overview on the fundamentals of neural networks and their technical details. Through three examples, we show their capabilities in the modelling of cosmological data, numerical tasks (saving computational time), and the classification of stellar objects. Artificial neural networks offer interesting qualities that make them viable alternatives for data analysis in cosmological research.

show abstract

What can machine learning tell us about the background expansion of the Universe?

Cited by 62 publications

References 89 publications

Nonparametric reconstruction of interaction in the cosmic dark sector

Nonparametric reconstruction of interaction in the cosmic dark sector

Euclid: Forecast constraints on consistency tests of the $Λ$CDM model

Observational Cosmology with Artificial Neural Networks

Contact Info

Product

Resources

About