How to use GP: Effects of the mean function and hyperparameter selection on Gaussian Process regression

Abstract

Gaussian processes have been widely used in cosmology to reconstruct cosmological quantities in a model-independent way. However, the validity of the adopted mean function and hyperparameters, and the dependence of the results on the choice have not been well explored. In this paper, we study the effects of the underlying mean function and the hyperparameter selection on the reconstruction of the distance moduli from type Ia supernovae. We show that the choice of an arbitrary mean function affects the reconstruction: a zero mean function leads to unphysical distance moduli and the best-fit LCDM to biased reconstructions. We propose to marginalize over a family of mean functions and over the hyperparameters to effectively remove their impact on the reconstructions. We further explore the validity and consistency of the results considering different kernel functions and show that our method is unbiased.

arXiv comment18 pages, 11 figures

References in corpus (24)

• Observational Evidence from Supernovae for an Accelerating Universe and a Cosmological Constant
• Measurements of Omega and Lambda from 42 High-Redshift Supernovae
• emcee: The MCMC Hammer
• The Complete Light-curve Sample of Spectroscopically Confirmed Type Ia Supernovae from Pan-STARRS1 and Cosmological Constraints from The Combined Pantheon Sample
• Exploring the Expansion History of the Universe
• Accelerating Universes with Scaling Dark Matter
• A Comprehensive Measurement of the Local Value of the Hubble Constant with 1 km/s/Mpc Uncertainty from the Hubble Space Telescope and the SH0ES Team
• Milky Way Cepheid Standards for Measuring Cosmic Distances and Application to Gaia DR2: Implications for the Hubble Constant
• Reconstruction of dark energy and expansion dynamics using Gaussian processes
• Gaussian Process Cosmography
• Nonparametric Dark Energy Reconstruction from Supernova Data
• A Simple Phenomenological Emergent Dark Energy Model can Resolve the Hubble Tension
• Smoothing Supernova Data to Reconstruct the Expansion History of the Universe and its Age
• Nonparametric Reconstruction of the Dark Energy Equation of State
• Cosmic Calibration: Constraints from the Matter Power Spectrum and the Cosmic Microwave Background
• Model Independent Reconstruction of the Expansion History of the Universe and the Properties of Dark Energy
• Nonparametric Reconstruction of the Dark Energy Equation of State from Diverse Data Sets
• Falsifying $\Lambda$CDM: Model-independent tests of the concordance model with eBOSS DR14Q and Pantheon
• Model independent inference of the expansion history and implications for the growth of structure
• Model-independent constraints on ${\Omega }_m$ and $H(z)$ from the link between geometry and growth
• Kernel-Based Emulator for the 3D Matter Power Spectrum from CLASS
• Debiasing Cosmic Gravitational Wave Sirens
• Defying the laws of Gravity I: model-independent reconstruction of the Universe expansion from growth data
• Gaussian Process Reconstruction of Reionization History

Cited by in corpus (34)

• The CosmoVerse White Paper: Addressing observational tensions in cosmology with systematics and fundamental physics
• Nonparametric late-time expansion history reconstruction and implications for the Hubble tension in light of recent DESI and type Ia supernovae data
• Cosmic chronometers to calibrate the ladders and measure the curvature of the Universe. A model-independent study
• Model-agnostic assessment of dark energy after DESI DR1 BAO
• Model-independent cosmological inference post DESI DR1 BAO measurements
• Model independent bounds on Type Ia supernova absolute peak magnitude
• On the consistency of $\Lambda$CDM with CMB measurements in light of the latest Planck, ACT, and SPT data
• Kernel Selection for Gaussian Process in Cosmology: with Approximate Bayesian Computation Rejection and Nested Sampling
• Exploring new physics in the late Universe's expansion through non-parametric inference
• A thorough investigation of the prospects of eLISA in addressing the Hubble tension: Fisher Forecast, MCMC and Machine Learning
• A Model-independent Method to Determine $H_0$ Using Time-Delay Lensing, Quasars and Type Ia Supernovae
• Non-parametric reconstruction of cosmological observables using Gaussian Processes Regression
• LADDER: Revisiting the Cosmic Distance Ladder with Deep Learning Approaches and Exploring its Applications
• Determination of cosmic curvature independent of the sound horizon and $H_0$ using BOSS/eBOSS and DESI DR1 BAO observations
• Testing Cosmology with Double Source Lensing
• Reconstructing the growth index $\gamma$ with Gaussian Processes
• A possible late-time transition of $M_B$ inferred via neural networks
• Determining Cosmological-model-independent $H_0$ with Gravitationally Lensed Supernova Refsdal
• Revisiting model-independent constraints on spatial curvature and cosmic ladders calibration: updated and forecast analyses
• Litmus tests of the flat $\Lambda$CDM model and model-independent measurement of $H_0{r}_{\mathrm{d}}$ with LSST and DESI
• A comprehensive data-driven odyssey to explore the equation of state of dark energy
• Is ${\omega }_0{\omega }_a$CDM a good model for the clumpy Universe?
• Dark Energy in the DESI Era: A Brief Review of Evidence, Beyond-$\Lambda$CDM Interpretations, and Tensions
• Redshift evolution of the X-ray and UV luminosity relation of quasars: calibrated results from SNe Ia
• Probing the Cosmic Distance Duality Relation via Non-Parametric Reconstruction for High Redshifts
• When Dark Matter Heats Up: A Model-Independent Search for Non-Cold Behavior
• Model-independent calibration of Gamma-Ray Bursts with neural networks
• Testing the Constancy of Type Ia Supernova Luminosities with Gaussian Process
• Constraining the primordial power spectrum using a differentiable likelihood
• Model independent calibration for sound horizon combining observations of supernovae and transversal BAO measurements
• Revisiting the Hubble Tension with DESI DR2 Baryon Acoustic Oscillation Observations and Machine Learning Methods
• EFT of Dark Energy with Cosmic Chronometers: Reconstructing Background EFT Functions
• Probing the Hubble Tension with an Infinite-Future Condition on the Hubble Parameter
• Breaking the Mass-sheet Degeneracy in Time-delay Cosmology with Lensed and Unlensed Type Ia Supernovae