Learning to Inflate

Abstract

Motivated by machine learning, we introduce a novel method for randomly generating inflationary potentials. Namely, we treat the Taylor coefficients of the potential as weights in a single-layer neural network and use gradient ascent to maximize the number of e-folds of inflation. Inflationary potentials "learned" in this way develop a critical point, which is typically a local maximum but may also be an inflection point. We study the phenomenology of the models along the gradient ascent trajectory, finding substantial agreement with experiment for large-field local maximum models and small-field inflection point models. For two-field models of inflation, the potential eventually learns a genuine multi-field model in which the inflaton curves significantly during the course of its descent.

31 pages, 12 figures, references added v2

References in corpus (26)

• Classification without labels: Learning from mixed samples in high energy physics
• Searching for New Physics with Deep Autoencoders
• Anomaly Detection for Resonant New Physics with Machine Learning
• A generic anti-QCD jet tagger
• Constraining Effective Field Theories with Machine Learning
• Fitting the Gamma-Ray Spectrum from Dark Matter with DMFIT: GLAST and the Galactic Center Region
• On the Possibility of Large Axion Moduli Spaces
• Pileup Mitigation with Machine Learning (PUMML)
• Machine Learning of Calabi-Yau Volumes
• Machine Learning in the String Landscape
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• Evolving neural networks with genetic algorithms to study the String Landscape
• Machine Learning CICY Threefolds
• Accidental Inflation in String Theory
• Search for beautiful tetraquarks in the ${\rm Y}(1S)\mu \mu$ invariant-mass spectrum
• Expansion of All Multitrace Tree Level EYM Amplitudes
• Exploring a string-like landscape
• GANs for generating EFT models
• On holographic entanglement entropy of Horndeski black holes
• Inflation in Multi-field Modified DBM Potentials
• Learning non-Higgsable gauge groups in 4D F-theory
• Inflation on a Slippery Slope
• Random Functions via Dyson Brownian Motion: Progress and Problems
• Inflation as an Information Bottleneck - A strategy for identifying universality classes and making robust predictions
• Testing Two-Field Inflation
• Machine Learning Cosmic Expansion History

Cited by in corpus (10)

• Branes with Brains: Exploring String Vacua with Deep Reinforcement Learning
• Conditions for (No) Eternal Inflation
• Machine learning Calabi-Yau metrics
• Statistical Predictions in String Theory and Deep Generative Models
• Machine learning cosmic inflation
• Cosmic Inflation and Genetic Algorithms
• Optimal Inflationary Potentials
• A Triumvirate of AI Driven Theoretical Discovery
• Canonical Scalar Field Inflation with a Woods-Saxon Potential
• Universes as Big Data