Cosmo-PINN: A Physics-Informed Neural Network for Cosmological Reconstruction

Andronikos Paliathanasis · 2026

Read Cosmo-PINN: A Physics-Informed Neural Network for Cosmological Reconstruction by Andronikos Paliathanasis (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-30139.txt.

Introduction

Andronikos Paliathanasis introduces Cosmo-PINN, a physics-informed network that rebuilds the dark energy equation of state from late-time observations. Standard Gaussian process and neural rebuilds follow data but need not respect physics. Cosmo-PINN embeds constraints into its loss function so rebuilt quantities obey physical laws. Training uses DESI baryon acoustic oscillation data, cosmic chronometers and supernova compilations. The rebuilt equation of state crosses the phantom divide around redshift one half, matching the Chevallier-Polarski-Linder model, and hits the standard limit in the past for quintessence.

Catalog record

CREATORAndronikos Paliathanasis
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE46641
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.30139

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Cosmo-PINN: A Physics-Informed Neural Network for Cosmological Reconstruction Andronikos Paliathanasis Abstract We introduce Cosmo-PINN, a Physics-Informed Neural Network for reconstruction of the cosmological theory. In this work we demonstrate the application of the Cosmo-PINN in the reconstruction of the dark energy equation of state parameter w D ​ E ​ ( z ) w_{DE}\left(z\right) directly from late-time cosmological observations. This framework overcomes the main limitation shared by Gaussian Process and Artificial Neural Network reconstruction approaches, where the recovered solution is driven by the data and it is not necessarily true that it is physically consistent, by embedding the cosmological constraints directly into the loss function as hard constraints, ensuring that the reconstructed quantities satisfy the physical laws at every point during the training. For the training of the network, we employed background data, and specifically the Baryon Acoustic Oscillation from DESI DR2, the Cosmic Chronometers and three different Supernova compilations, while we simultaneously introduce the cosmological parameters H 0 , Ω m ​ 0 H_{0},~\Omega_{m0} and r drag r_{\mathrm{drag}} as trained parameters. The reconstruction shows that the trained w D ​ E ​ ( z ) w_{DE}\left(z\right) crosses the phantom divide within the redshift range z = 0.27 − 0.42 z=0.27-0.42 in agreement with the value obtained by the Chevallier-Polarski-Linder model. In the quintessence scenario, for large redshifts the dark energy Ω D ​ E ​ ( z ) \Omega_{DE}\left(z\right) provides a pressureless nonzero contribution to the cosmological fluid suggesting a unified scenario. Finally, we demonstrate the significance of imposing the physical constraints within the loss function by comparing the Cosmo-PINN reconstruction against a purely data-driven neural network with the same architecture. I Introduction The mechanism responsible for the late-time acceleration phase of the observable universe rr1 ; Teg ; Kowal ; Komatsu ; suzuki11 remains one of the central open problems in modern cosmology. The gravitational field describes attractive interactions, however in order to describe the late-time acceleration in the framework of General Relativity, it is required to introduce a matter source with repulsive force contribution in order to drive the cosmic acceleration. This component, which is detected only through its gravitational effects, is referred as dark energy jo ; jo1 . The cosmological constant is the simplest theoretical motivated dark energy candidate. The cosmological constant introduces the minimum degrees of freedom in the gravitational field equations and until recently it was the standard model for the description of the cosmic acceleration. The corresponding cosmological model, the \Lambda CDM, is nowadays challenged on the observational side by the cosmological tensions ten2 . Furthermore, the cosmological constant suffers from two other major problems, the fine tuning and the coincidence problems Weinberg89 ; Pad03 ; Peri08 . A plethora of alternative proposals to the cosmological constant have been introduced in the literature in order to address the above issues, we refer the reader to ter3 for a systematic review. The proposed solutions can be classified into models which modify the matter sector, such as Chaplygin gas models cg1 , interacting dark sector int1 ; int2 ; int3 , scalar fields ratra ; peebles ; pha ; mp1 ; mp2 ; mp3 ; luongo , k -essence kes , and modifications of the gravitational action integral, including scalar-tensor models st , Horndeski theory hr , f\left(R\right) -gravity fr , teleparallel f\left(T\right) -gravity ft , symmetric teleparallel f\left(Q\right) -gravity fq and many others md1 ; md2 ; md3 ; md4 ; md5 ; md6 ; md7 ; md8 ; md9 ; md10 . Phenomenological approaches form an alternative and simpler framework for the description of the cosmological data, which allows us to extract important information about the nature of dark energy, since they allow us to explore how dark energy is distinct from the cosmological constant scenario cpl2 ; cpl4 . The most widely used phenomenological dark energy model is the Chevallier-Polarski-Linder (CPL) model cpl1 ; cpl1a . CPL belongs to the class of w_{0}w_{a} CDM models, in which the dark energy equation of state parameter is considered to be a function of the redshift with two free variables, where w_{0} usually refers to the value of the dark energy equation of state parameter at the present, for more details we refer the reader to par1 ; par4 ; par5 ; exp1A ; osc1 ; osc3 ; osc5 and references therein. Nevertheless, phenomenological parametrizations are model-dependent because the functional form of the w_{DE}\left(z\right) is imposed a priori, and the resulting observable predictions therefore carry the imprint of the assumed function rather than of the underlying physics. A reconstruction approach is essential in order to extract information for the properties of the physical theory, directly from the observational data, without restricting the free functional forms of the theory beforehand. In the standard Bayesian method, the parameter estimation is usually performed under a specific assumption for the cosmological model, that is a given functional form of the dark energy equation of state w_{DE}\left(z\right) , or the Hubble function H\left(z\right)\, , such as the \Lambda CDM model or the CPL parametrization discussed above. The limitation of the Bayesian approach is that the conclusions made are valid only for the specific model assumed. Various phenomenological models for the dark energy equation of state can exhibit similar behaviour at a given cosmological epoch and fit the observational data in a similar way, such that the models are statistically indistinguishable anlcdm . Moreover, in theoretically motivated models, such as the quintessence scalar field, there exists a plethora of scalar field potentials that share a common attractor solution, causing all such models to predict a similar dynamical evolution map1 which as a result leads to a similar fitting to the background data. Two main approaches to the reconstruction problem which can be found in the literature are the Gaussian Process (GP) gp1 ; gp2 ; gp3 ; gp4 ; gp5 ; gp6 ; gp7 ; gp8 ; gp9 ; gp10 ; gp11 ; gp12 ; gp13 ; gp14 ; gp15 and the Artificial Neural Networks (ANNs) an1 ; an2 ; an3 ; an4 ; an5a ; an5 ; an6 ; an7 ; an8 ; an9 . GP is a model-independent approach, however, it suffers from two main issues. The selection of the kernel may affect the reconstructed physical parameters gp16a ; gp16 , while the GP suffers from overfitting issues and it is sensitive to the H_{0} value. On the other hand, the ANN approach provides a fully data-driven, model-independent reconstruction framework; however, there is no guarantee that the trained solution satisfies the field equations governing the cosmological dynamics, since the physical laws are not incorporated into the training. Physics-Informed Neural Networks (PINNs) pinnp , in which the physical laws, that is, the field equations in cosmological studies, are embedded directly into the loss function offer a natural approach to overcome this limitation during the reconstruction pinnbook . PINNs have been applied previously in cosmological studies. In pnn1 a PINN was used to solve the background cosmological field equations both in General Relativity and in modified theories of gravity. Furthermore, in pnn2 ; pnn3 , the PINN approach was employed to reconstruct the Hubble function within the Tsallis and Barrow holographic dark energy models, where the field equations were embedded

Read the complete text