The best inflationary models after Planck

Jérôme Martin, Christophe Ringeval, Roberto Trotta, Vincent VenninJournal of Cosmology and Astroparticle Physics 2014 · DOI · arXiv

Abstract

We compute the Bayesian evidence and complexity of 193 slow-roll single-field models of inflation using the Planck 2013 Cosmic Microwave Background data, with the aim of establishing which models are favoured from a Bayesian perspective. Our calculations employ a new numerical pipeline interfacing an inflationary effective likelihood with the slow-roll library ASPIC and the nested sampling algorithm MultiNest. The models considered represent a complete and systematic scan of the entire landscape of inflationary scenarios proposed so far. Our analysis singles out the most probable models (from an Occam's razor point of view) that are compatible with Planck data, while ruling out with very strong evidence 34% of the models considered. We identify 26% of the models that are favoured by the Bayesian evidence, corresponding to 15 different potential shapes. If the Bayesian complexity is included in the analysis, only 9% of the models are preferred, corresponding to only 9 different potential shapes. These shapes are all of the plateau type.

Introduction

Cosmic inflation is the leading paradigm describing the exponential expansion of the universe in its first fractions of a second. This expansion explains the universe's flatness and uniformity while providing the quantum fluctuations that seeded all large-scale structures. However, theorists have proposed hundreds of different "potentials"—mathematical shapes describing how inflation starts and ends. This has led to a "landscape problem" where there are too many theories and insufficient means to distinguish them. This study addresses this by performing a systematic "statistical audit" of 193 single-field inflationary models using high-precision data from the Planck satellite.

Methodology

The researchers developed an automated numerical pipeline to evaluate the inflationary landscape using three core tools:

  1. ASPIC: A library containing the physics and "slow-roll" parameters for 193 different inflationary scenarios.
  2. MultiNest: A nested sampling algorithm used to calculate Bayesian evidence. This value represents how well a model fits the data while naturally penalizing overly complex models—a mathematical application of Occam’s Razor.
  3. Effective Likelihood: A computational framework that marginalizes over post-inflationary parameters (like the details of the "reheating" phase), allowing a pure focus on the inflationary potential.

Beyond simple fits, the study calculated Bayesian complexity, which measures how many parameters a model actually utilizes to explain the data. The primary dataset used was the Planck 2013 Cosmic Microwave Background (CMB) data, supplemented by WMAP polarization data and Baryon Acoustic Oscillations (BAO).

Results

The core of the analysis involves ranking models based on the log-Bayesian evidence ratio ($|\ln B^i_{\text{REF}}|$). The strength of the evidence for or against a model is interpreted using a quantitative scale.

Jeffreys' scale for the strength of Bayesian evidence.

Using this rigorous framework, the study yielded several key insights:

  • Model Pruning: 34% of the models were ruled out with "very strong" evidence.
  • Favoured Models: 26% of the models (representing 15 potential shapes) were found to be favoured or inconclusive relative to the best-performing model.
  • Complexity Filter: When Bayesian complexity was factored in, only 9% of the models (9 potential shapes) remained preferred.
  • Plateau Supremacy: The surviving models are almost exclusively "plateau-type" potentials, where the energy density flattens at large field values. This includes the Starobinsky model and Higgs Inflation.
  • Top Performer: The analysis identifies Kähler Moduli Inflation II (KMIII) as the most probable model, though it is statistically comparable to Higgs Inflation.

Discussion and Conclusion

The study demonstrates that many complex inflationary models are "over-engineered." Bayesian evidence favors simplicity, and the Planck data strongly points toward plateau-type potentials that match the observed "spectral tilt" ($n_s \approx 0.96$).

However, limitations remain. The study is restricted to single-field models and relies on the 2013 Planck data release. Future data, particularly regarding B-mode polarization, will be necessary to further distinguish between the remaining plateau models. Furthermore, the transition from inflation to the hot Big Bang (reheating) remains a significant source of uncertainty that must be accounted for in future theoretical frameworks.

Summary updated

Abstract

We compute the Bayesian evidence and complexity of 193 slow-roll single-field models of inflation using the Planck 2013 Cosmic Microwave Background data, with the aim of establishing which models are favoured from a Bayesian perspective. Our calculations employ a new numerical pipeline interfacing an inflationary effective likelihood with the slow-roll library ASPIC and the nested sampling algorithm MultiNest. The models considered represent a complete and systematic scan of the entire landscape of inflationary scenarios proposed so far. Our analysis singles out the most probable models (from an Occam's razor point of view) that are compatible with Planck data, while ruling out with very strong evidence 34% of the models considered. We identify 26% of the models that are favoured by the Bayesian evidence, corresponding to 15 different potential shapes. If the Bayesian complexity is included in the analysis, only 9% of the models are preferred, corresponding to only 9 different potential shapes. These shapes are all of the plateau type. © 2014 IOP Publishing Ltd and Sissa Medialab srl.
BibTeX
@article{martin2014best,
  author = {Jérôme Martin and Christophe Ringeval and Roberto Trotta and Vincent Vennin},
  title = {{The best inflationary models after Planck}},
  journal = {Journal of Cosmology and Astroparticle Physics},
  year = {2014},
  doi = {10.1088/1475-7516/2014/03/039},
}
Figure