Learning to Inflate
Author:
Preprint
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.Keywords
All Related Versions
This publication has 54 references indexed in Scilit:
- Learning non-Higgsable gauge groups in 4D F-theoryJournal of High Energy Physics, 2018
- Evolving neural networks with genetic algorithms to study the string landscapeJournal of High Energy Physics, 2017
- Weakly supervised classification in high energy physicsJournal of High Energy Physics, 2017
- Inflation with a graceful exit in a random landscapeJournal of High Energy Physics, 2017
- Systematics of axion inflation in Calabi-Yau hypersurfacesJournal of High Energy Physics, 2017
- Deep learning in color: towards automated quark/gluon jet discriminationJournal of High Energy Physics, 2017
- Jet-images — deep learning editionJournal of High Energy Physics, 2016
- Axion decay constants away from the lamppostJournal of High Energy Physics, 2016
- Transplanckian axions!?Journal of High Energy Physics, 2015
- On Gaussian random supergravityJournal of High Energy Physics, 2014