arXiv subjects

Suresh Kumar

Publications and source records attributed to Suresh Kumar.

At least 19 recordsLinked to original sources

arXiv2026

Probing late-time deviations from $\Lambda$CDM with a quadratic dark energy expansion

We investigate the observational viability of a quadratic dark energy expansion (QDEE) model as a phenomenological extension of the standard $\Lambda\mathrm{CDM}$ cosmological framework. This approach introduces the additional degrees of freedom that permit mild late-time deviations from a constant dark-energy component while preserving the standard early-Universe behavior. We constrain the model using a comprehensive combination of cosmological datasets, including Planck 2018 cosmic microwave background (CMB) measurements, Atacama Cosmology Telescope (ACT) Data Release 6 (DR6) and South Pole Telescope (SPT-3G) data, Dark Energy Spectroscopic Instrument (DESI) Data Release 2 (DR2), and the Pantheon Plus type Ia supernova compilation. Our results show that the QDEE framework shifts the inferred Hubble constant toward higher values relative to $\Lambda\mathrm{CDM}$, partially alleviating the tension with local measurements while remaining consistent with early-Universe constraints. Bayesian model comparison indicates strong evidence in favor of this framework over standard $\Lambda\mathrm{CDM}$ across multiple dataset combinations. Posterior predictive checks further demonstrate that the model yields predictions consistent with the observed data within statistical uncertainties.

astro-ph.CO↗

arXiv2026

Statistical consistency of sign-switching vacuum energy with cosmological observations

We assess dataset agreement and late-time predictive adequacy in $\Lambda$CDM and its sign-switching extension, $\Lambda_{\rm s}$CDM, using a suite of Gaussian and exact non-Gaussian consistency diagnostics. Both models are constrained with cosmic microwave background measurements from Planck, ACT, and SPT, baryon acoustic oscillation data from DESI DR2, and low-redshift Type Ia supernova data from PantheonPlus+SH0ES. We find that commonly used Gaussian tension metrics can significantly overstate inconsistencies when broad, non-Gaussian posteriors are combined with tightly constrained datasets. In contrast, the exact non-Gaussian parameter shift indicates excellent consistency between CMB and BAO observations in both models. The $\Lambda_{\rm s}$CDM extension modestly improves geometric compatibility at intermediate redshifts, although reductions in parameter-level tension do not necessarily imply improved predictive consistency. These results highlight the importance of exact, non-Gaussian, and predictive diagnostics for robust assessments of cosmological model consistency.

astro-ph.CO↗

arXiv2026

Probing dynamical dark energy with late-time data: Evidence, tensions, and the limits of the $w_0w_a$CDM framework

We test the dynamical dark-energy $w_0w_a$CDM (CPL) framework against $\Lambda$CDM using CMB anisotropies and lensing together with DESI DR2, SDSS-IV, transverse/angular BAO (BAOtr), and Cepheid-calibrated PantheonPlus SN~Ia data. CPL inferences are strongly dataset-dependent. CMB data alone leave a broad geometric degeneracy, while DESI DR2 BAO pulls the reconstruction toward weak present-day acceleration. In contrast, CMB combined with PP\&SH0ES and BAOtr favors a moderately accelerating expansion and substantially reduces the Hubble tension. The origin of this behavior can be traced to low-redshift distance information: BAOtr and DESI/SDSS prefer different BAO distance ratios at $z\lesssim0.5$, which drives divergent CPL expansion histories. We quantify this mismatch directly at the data level by comparing angular BAO scales, including the common $z=0.510$ point and a conservative local interpolation of BAOtr with no extrapolation. As expected within CPL, where pre-recombination physics is fixed, $r_{\rm d}$ remains nearly unchanged, so shifts in $H_0$ are absorbed by late-time expansion freedom rather than by a change in the sound horizon. Bayesian evidence is likewise contingent on the low-redshift data: it favors CPL mainly when PP\&SH0ES and/or BAOtr are included, is inconclusive for CMB-only and CMB+DESI, and moderately favors $\Lambda$CDM for CMB+SDSS. These results show that apparent support for CPL and its ability to ease the Hubble tension are not universal, motivating more flexible late-time models and closer scrutiny of BAO systematics.

astro-ph.CO↗

arXiv2025

$\mathbf{\Omega_1\Omega_2}$-$\mathbf{\Lambda}$CDM: A promising phenomenological extension of the standard model of cosmology

We investigate a phenomenological extension of the standard $\Lambda$CDM framework, the $\Omega_1\Omega_2$-$\Lambda$CDM model, in which the total energy density of the universe is expanded in powers of $1+z$. This parameterization recovers $\Lambda$CDM and introduces two additional terms, $\Omega_1(1+z)$ and $\Omega_2(1+z)^2$, in the dark energy sector alongside the cosmological constant, leading to a physically interpretable and observationally testable effective dark energy dynamics. Using Planck cosmic microwave background (CMB) data, we find that the model allows additional freedom in the late-time expansion history while preserving standard early-universe physics. When DESI baryon acoustic oscillation (BAO) data are included, the inferred Hubble constant is $H_0 = 69.74 \pm 0.77~\mathrm{km\;s^{-1}\;Mpc^{-1}}$ (68% CL), consistent with Tip of the Red Giant Branch (TRGB) measurements from the Carnegie--Chicago Hubble Program (CCHP). The reconstructed dark energy equation of state exhibits a smooth transition across the phantom divide followed by asymptotic de Sitter behavior, modifying late-time dynamics while maintaining standard early-time cosmology. Overall, controlled late-time deviations from $\Lambda$CDM improve cosmological concordance.

astro-ph.CO↗

arXiv2025

New constraints on interacting dark energy from DESI DR2 BAO observations

In its second data release (DR2), the Dark Energy Spectroscopic Instrument (DESI) publicly released measurements of Baryon Acoustic Oscillations (BAO) from over 13.1 million galaxies and 1.6 million quasars, covering the redshift range $0.295 \leq z \leq 2.330$. In this work, we investigate the impact of this new dataset on dark sector interaction models, which are motivated by non-gravitational interactions between dark energy (DE) and dark matter (DM), commonly referred to as interacting dark energy models (IDE). We focus on two frameworks: the traditional IDE model and the recently proposed sign-switching Interacting model (S-IDE), aiming to derive new and robust constraints on both scenarios. After carefully selecting the sample for the joint analysis, ensuring compatibility among the data without significant tension, our main results indicate that both models can alleviate the $H_0$ tension, reducing it to moderate tension approximately $2.7\sigma$. The IDE model shows compatibility with the latest $S_8$ constraints from cosmic shear surveys, while the S-IDE model predicts lower values of $S_8$, which align with alternative perspectives on the $S_8$ tension. For the traditional IDE model, we derive new bounds for the coupling parameter, marking the strongest constraints to date through geometric measurements. This highlights the crucial role that supernova samples can play in refining these constraints. For the S-IDE model, we find mild evidence (over $2\sigma$) for a non-zero coupling, once the PantheonPlus dataset calibrated with Cepheid-based magnitude measurements is included in the analysis.

astro-ph.CO↗

arXiv2025

Artificial Intelligence-Driven Prognostic Classification of COVID-19 Using Chest X-rays: A Deep Learning Approach

Background: The COVID-19 pandemic has overwhelmed healthcare systems, emphasizing the need for AI-driven tools to assist in rapid and accurate patient prognosis. Chest X-ray imaging is a widely available diagnostic tool, but existing methods for prognosis classification lack scalability and efficiency. Objective: This study presents a high-accuracy deep learning model for classifying COVID-19 severity (Mild, Moderate, and Severe) using Chest X-ray images, developed on Microsoft Azure Custom Vision. Methods: Using a dataset of 1,103 confirmed COVID-19 X-ray images from AIforCOVID, we trained and validated a deep learning model leveraging Convolutional Neural Networks (CNNs). The model was evaluated on an unseen dataset to measure accuracy, precision, and recall. Results: Our model achieved an average accuracy of 97%, with specificity of 99%, sensitivity of 87%, and an F1-score of 93.11%. When classifying COVID-19 severity, the model achieved accuracies of 89.03% (Mild), 95.77% (Moderate), and 81.16% (Severe). These results demonstrate the model's potential for real-world clinical applications, aiding in faster decision-making and improved resource allocation. Conclusion: AI-driven prognosis classification using deep learning can significantly enhance COVID-19 patient management, enabling early intervention and efficient triaging. Our study provides a scalable, high-accuracy AI framework for integrating deep learning into routine clinical workflows. Future work should focus on expanding datasets, external validation, and regulatory compliance to facilitate clinical adoption.

eess.IV↗

arXiv2025

Mapping the $\Lambda_{\rm s}$CDM scenario to $f(T)$ modified gravity: Effects on structure growth rate

The concept of a rapidly sign-switching cosmological constant, interpreted as a mirror AdS-dS transition in the late universe and known as the $\Lambda_{\rm s}$CDM, has significantly improved the fit to observational data, offering a promising framework for alleviating major cosmological tensions such as the $H_0$ and $S_8$ tensions. However, when considered within general relativity, this scenario does not predict any effects on the evolution of the matter density contrast beyond modifications to the background functions. In this work, we propose a new gravitational model in which the background dynamics predicted by the $\Lambda_{\rm s}$CDM framework are mapped into $f(T)$ gravity, dubbed $f(T)-\Lambda_{\rm s}$CDM, rendering the models indistinguishable at the background level. However, in this new scenario, the sign-switching cosmological constant dynamics modify the evolution of linear matter perturbations through an effective gravitational constant, $G_{\rm eff}$. We investigate the evolution of the growth rate and derive new observational constraints for this scenario using RSD measurements. We also present new constraints in the standard $\Lambda_{\rm s}$CDM case, incorporating the latest Type Ia supernovae data samples available in the literature, along with BAO data from DESI. Our findings indicate that the new corrections expected at the linear perturbative level, as revealed through RSD samples, can provide significant evidence in favor of this new scenario. Additionally, this model may be an excellent candidate for resolving the current $S_8$ tension.

astro-ph.CO↗

arXiv2025

Sign Switching in Dark Sector Coupling Interactions as a Candidate for Resolving Cosmological Tensions

The $\Lambda$CDM model has successfully explained a wide range of cosmological observations, but is increasingly challenged by the emergence of cosmological tensions, particularly the Hubble Tension $H_0$ and the $S_8$ tension. The Hubble Tension, with a significance above 5$\sigma$, and the $S_8$ tension, showing a discrepancy of approximately 2-4$\sigma$, highlight inconsistencies between measurements of the local and early universe. This paper expands a well-established Interacting Dark Energy (IDE) phenomenological scenario, where dark matter (DM) can transfer energy to dark energy (DE) or vice versa, depending on the sign of the coupling parameter $\xi$. The novel feature consists in a transition mechanism which reverses the direction of the energy-momentum transfer after the redshift where the densities of the dark species are the same. We evaluate this model using a comprehensive set of recent observational data, including Baryon Acoustic Oscillations (BAO) from the DESI survey, Type Ia Supernovae from the PantheonPlus, DESY5 and Union3 samples, and Cosmic Microwave Background (CMB) data from Planck. Our analysis shows that this scenario can potentially relax both the $H_0$ and $S_8$ tensions simultaneously. We find the new model to be weakly preferred over $\Lambda$CDM by BAO-DESI data. However, we show that the IDE model features positive Bayesian evidence compared to $\Lambda$CDM only when Cepheid distance calibration in the SH0ES sample is used to calibrate SNIa data from PantheonPlus.

astro-ph.CO↗

arXiv2024

Quantifying the $S_8$ tension and evidence for interacting dark energy from redshift-space distortion measurements

In recent years, Cosmic Microwave Background (CMB) observations, Weak Lensing surveys, and $f\sigma_8(z)$ measurements from Redshift-Space Distortions (RSD) have revealed a significant ($\sim$3$-$5$\sigma$) discrepancy in the inferred value of the matter clustering parameter $S_8$. In this work, we investigate the implications of RSD for a cosmological framework postulating an interaction between Dark Energy (DE) and Dark Matter (DM). We explore scenarios where DM can transfer energy-momentum to DE or vice versa. The energy-momentum flow is characterized by the strength and the sign of the coupling parameter $\xi$. Our baseline analysis combines RSD measurements with the latest data from Baryon Acoustic Oscillations (BAO) observed by DESI, Type Ia Supernovae from the PantheonPlus sample, and CMB data from Planck. We demonstrate that RSD measurements provide significant additional information imposing new and strong upper bounds on possible interaction in the dark sector. Models with $\xi > 0$ can effectively alleviate the tension in $S_8$, presenting them as compelling alternatives.

astro-ph.CO↗

arXiv2024

$\Lambda_{\rm s}$CDM cosmology: Alleviating major cosmological tensions by predicting standard neutrino properties

We investigate a two-parameter extension of the $\Lambda_{\rm s}$CDM model by allowing variations in the effective number of neutrino species $N_{\rm eff}$ and their total mass $\sum m_\nu$. Our motivation is twofold: (i) to examine whether $\Lambda_{\rm s}$CDM retains its success in fitting the data and addressing major cosmological tensions, without suggesting a need for a deviation from the standard model of particle physics, and (ii) to determine whether the data indicate new physics that could potentially address cosmological tensions, either in the post-recombination universe through the late-time mirror AdS-dS transition, or in the pre-recombination universe through modifications in the standard values of $N_{\rm eff}$ and $\sum m_\nu$, or both. Within the extended $\Lambda_{\rm s}$CDM model, referred to as $\Lambda_{\rm s}$CDM+$N_{\rm eff}$+$\sum m_{\rm \nu}$, we find no significant tension when considering the Planck-alone analysis. We observe that incorporating BAO data limits the further success of the $\Lambda_{\rm s}$CDM extension. However, the weakly model-dependent BAOtr data, along with Planck and Planck+PP\&SH0ES, favor $H_0\sim 73\,{\rm km\, s^{-1}\, Mpc^{-1}}$. In cases where BAOtr dataset is used, the mirror AdS-dS transition is very effective in providing enhanced $H_0$ values, and thus the model requires no significant deviation from the standard value of $N_{\rm eff} = 3.044$. Both the $H_0$ and $S_8$ tensions are effectively addressed, with some compromise in the case of the Planck+BAO dataset. Finally, the upper bounds obtained on $\sum m_\nu \lesssim 0.5$~eV are fully compatible with neutrino oscillation experiments. Our findings provide evidence that late-time physics beyond $\Lambda$CDM, such as $\Lambda_{\rm s}$CDM, without altering the standard pre-recombination universe, can suffice to alleviate the major cosmological tensions.

astro-ph.CO↗

arXiv2024

Cosmological constraints on $\Lambda_{\rm s}$CDM scenario in a type II minimally modified gravity

The idea of a rapid sign-switching cosmological constant (mirror AdS-dS transition) in the late universe at $z\sim1.7$, known as the $\Lambda_{\rm s}$CDM model, has significantly improved the fit to observational data and provides a promising scenario for alleviating major cosmological tensions, such as the $H_0$ and $S_8$ tensions. However, in the absence of a fully predictive model, implementing this fit required conjecturing that the dynamics of the linear perturbations are governed by general relativity. Recent work embedding the $\Lambda_{\rm s}$CDM model with the Lagrangian of a type II minimally modified gravity known as VCDM has propelled $\Lambda_{\rm s}$CDM to a fully predictive model, removing the uncertainty related to the aforementioned assumption; we call this new model $\Lambda_{\rm s}$VCDM. In this work, we demonstrate that not only does $\Lambda_{\rm s}$CDM fit the data better than the standard $\Lambda$CDM model, but the new model, $\Lambda_{\rm s}$VCDM, performs even better in alleviating cosmological tensions while also providing a better fit to the data, including CMB, BAO, SNe Ia, and cosmic shear measurements. Our findings highlight the $\Lambda_{\rm s}$CDM framework, particularly the $\Lambda_{\rm s}$VCDM model, as a compelling alternative to the standard $\Lambda$CDM model, especially by successfully alleviating the $H_0$ tension. Additionally, these models predict higher values for $\sigma_8$, indicating enhanced structuring, albeit with lower present-day matter density parameter values and consequently reduced $S_8$ values, alleviating the $S_8$ tension as well. This demonstrates that the data are well fit by a combination of background and linear perturbations, both having dynamics differing from those of $\Lambda$CDM. This paves the way for further exploration of new ways for embedding the sign-switching cosmological constant into other models.

astro-ph.CO↗

arXiv2024

$\Lambda_{\rm s}$CDM cosmology from a type-II minimally modified gravity

We integrate $\Lambda_{\rm s}$CDM, a promising scenario for alleviating cosmological tensions, into VCDM, a type-II minimally modified gravity. This promotes the scenario to a fully predictive model (dubbed $\Lambda_{\rm s}$VCDM) that specifies the cosmological evolution self-consistently, including through the late-time AdS-to-dS transition epoch. In this theory, an auxiliary scalar field generates an effective cosmological constant with either a constant or a linear potential. This allows an abrupt mirror AdS-to-dS transition via a piecewise-linear potential with a sudden slope change. To remove the associated sudden singularity and ensure stable evolution, we smooth the junction using a blended sigmoid interpolant, obtaining rapid but continuous transitions. We identify two qualitatively distinct smooth mirror AdS-to-dS realisations of $\Lambda_{\rm s}$: (i) an agitated transition, in which the potential interpolates between equal-magnitude AdS and dS plateaus and $\Lambda_{\rm s}$ develops a central bump; and (ii) a quiescent transition, in which the potential remains continuous but changes slope across the transition layer, so that $\Lambda_{\rm s}(a)$ can remain monotone, with possible shallow shoulders, and a central bump is not automatic. Depending on type and sharpness, a finite-width transition can induce a transient accelerated-expansion interval ($\ddot a>0$) around $z\sim 1.5-2$, in addition to present-day acceleration, and, if the background enters a region where $V_{,\phi\phi}>2/3$, a nested super-acceleration episode. These distinct transient histories can imprint signatures on background and perturbation evolution. Our construction enables a self-consistent observational assessment of smooth $\Lambda_{\rm s}$CDM realisations and motivates multi-probe analyses to test transition dynamics and reassess cosmological tensions.

astro-ph.CO↗

arXiv2024

Effects of type Ia supernovae absolute magnitude priors on the Hubble constant value

We systematically explore the influence of the prior of the peak absolute magnitude ($M$) of type Ia supernovae (SNe Ia) on the measurement of the Hubble constant ($H_0$) from SNe Ia observations. We consider five different data-motivated $M$ priors, representing varying levels of dispersion, and assume the spatially-flat $\Lambda$CDM cosmological model. Different $M$ priors lead to relative changes in the mean values of $H_0$ from 2% to 7%. Loose priors on $M$ yield $H_0$ estimates consistent with both the Planck 2018 result and the SH0ES result at the 68% confidence level. We also examine the potential impact of peculiar velocity subtraction on the value of $H_0$, and show that it is insignificant for the SNe Ia observations with redshift $z > 0.01$ used in our analyses. We also repeat the analysis in the cosmography model and find very similar results. This suggests that our results are robust and model independent.

astro-ph.CO↗

arXiv2023

Observational bounds on extended minimal theories of massive gravity: New limits on the graviton mass

In this work, we derive for the first time observational constraints on the extended Minimal Theory of Massive Gravity (eMTMG) framework in light of Planck-CMB data, geometrical measurements from Baryon Acoustic Oscillation (BAO), Type Ia supernovae from the recent Pantheon+ samples, and also using the auto and cross-correlations cosmic shear measurements from KIDS-1000 survey. Given the great freedom of dynamics choice for the theory, we consider an observationally motivated subclass in which the background evolution of the Universe goes through a transition from a (positive or negative) value of the effective cosmological constant to another value. From the statistical point of view, we did not find evidence of such a transition, i.e. deviation from the standard $\Lambda$CDM behavior, and from the joint analysis using Planck + BAO + Pantheon+ data, we constrain the graviton mass to $< 6.6 \times 10^{-34}$ eV at 95% CL. We use KIDS-1000 survey data to constrain the evolution of the scalar perturbations of the model and its limits for the growth of structure predicted by the eMTMG scenario. In this case, we find small evidence at 95% CL for a non-zero graviton mass. We interpret and discuss these results in light of the current tension on the $S_8$ parameter. We conclude that, within the subclass considered, the current data are only able to impose upper bounds on the eMTMG dynamics. Given its potentialities beyond the subclass, eMTMG can be classified as a good candidate for modified gravity, serving as a framework in which observational data can effectively constrain (or confirm) the graviton mass and deviations from the standard $\Lambda$CDM behavior.

astro-ph.CO↗

arXiv2023

Deep Learning in Computational Biology: Advancements, Challenges, and Future Outlook

Deep learning has become a powerful tool in computational biology, revolutionising the analysis and interpretation of biological data over time. In our article review, we delve into various aspects of deep learning in computational biology. Specifically, we examine its history, advantages, and challenges. Our focus is on two primary applications: DNA sequence classification and prediction, as well as protein structure prediction from sequence data. Additionally, we provide insights into the outlook for this field. To fully harness the potential of deep learning in computational biology, it is crucial to address the challenges that come with it. These challenges include the requirement for large, labelled datasets and the interpretability of deep learning models. The use of deep learning in the analysis of DNA sequences has brought about a significant transformation in the detection of genomic variants and the analysis of gene expression. This has greatly contributed to the advancement of personalised medicine and drug discovery. Convolutional neural networks (CNNs) have been shown to be highly accurate in predicting genetic variations and gene expression levels. Deep learning techniques are used for analysing epigenetic data, including DNA methylation and histone modifications. This provides valuable insights into metabolic conditions and gene regulation. The field of protein structure prediction has been significantly impacted by deep learning, which has enabled accurate determination of the three-dimensional shape of proteins and prediction of their interactions. The future of deep learning in computational biology looks promising. With the development of advanced deep learning models and interpretation techniques, there is potential to overcome current challenges and further our understanding of biological systems.

cs.LG↗

arXiv2023

$\Lambda_{\rm s}$CDM model: A promising scenario for alleviation of cosmological tensions

We present a comprehensive analysis of the $\Lambda_{\rm s}$CDM model, which explores the recent conjecture suggesting a rapid transition of the Universe from anti-de Sitter vacua to de Sitter vacua (viz., the cosmological constant switches sign from negative to positive) at redshift ${z_\dagger\sim 2}$, inspired by the graduated dark energy (gDE) model. Our analysis shows that, predicting $z_\dagger\approx1.7$, $\Lambda_{\rm s}$CDM simultaneously addresses the major cosmological tensions of the standard $\Lambda$CDM model, viz., the Hubble constant $H_0$, the Type Ia Supernovae absolute magnitude $M_{\rm B}$, and the growth parameter $S_8$ tensions, along with other less significant tensions such as the BAO Lyman-$\alpha$ discrepancy.

astro-ph.CO↗

arXiv2023

New late-time constraints on $f(R)$ gravity

Modification of general relativity (GR) inspired by theories like $f(R)$ gravity is among the most popular ones to explain the late-time acceleration of the Universe as an alternative to the $\Lambda$CDM model. In this work, we use the state-of-the-art BAO+BBN data and the most recent Type Ia supernovae (SNe Ia) sample namely PantheonPlus, including the Cepheid host distances and covariance from SH0ES samples, to robustly constrain the $f(R)$ gravity framework via two of the most popular $f(R)$ models in literature, namely, the Hu-Sawicki and Starobinsky models. Additionally, we consider how the time variation of the Newton's gravitational constant affects the supernovae distance modulus relation. We find a minor evidence for $f(R)$ gravity under the Hu-Sawicki dynamics from BAO+BBN and BAO+BBN+uncalibrated supernovae joint analysis, but the inclusion of Cepheid host distances, makes the model compatible with GR. Further, we notice tendency of this model to relax the $H_0$ tension. In general, in all the analyses carried out in this study with the late time probes, we find both the $f(R)$ models to be consistent with GR at 95\% CL.

astro-ph.CO↗

arXiv2023

Exploring the $H_0$ tension and the evidence of dark sector interaction from 2D BAO measurements

We explore observational constraints on a cosmological model with an interaction between dark energy (DE) and dark matter (DM), using a compilation of 15 measurements of the 2D BAO (i.e., transversal) scale in combination with Planck-CMB data, to explore the parametric space of a class of interacting DE models. We find that 2D BAO measurements can generate different observational constraints compared to the traditional approach of studying the matter clustering in the 3D BAO measurements. Contrary to the observations for the $\Lambda$CDM and IDE models when analyzed with Planck-CMB + 3D BAO data, we note that Planck-CMB + 2D BAO data favor high values of the Hubble constant $H_0$. From the joint analysis with Planck-CMB + 2D BAO + Gaussian prior on $H_0$, we find $H_0 = 73.4 \pm 0.88$ km/s/Mpc. We conclude that the $H_0$ tension is solved in the IDE model with strong statistical evidence (more than 3$\sigma$) for the IDE cosmologies.

astro-ph.CO↗