An autoencoder for heterotic orbifolds with arbitrary geometry

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Escalante-Notario, Enrique, Portillo-Castillo, Ignacio, Ramos-Sanchez, Saul
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911767053991936
author Escalante-Notario, Enrique
Portillo-Castillo, Ignacio
Ramos-Sanchez, Saul
author_facet Escalante-Notario, Enrique
Portillo-Castillo, Ignacio
Ramos-Sanchez, Saul
contents Artificial neural networks have become important to improve the search for admissible string compactifications and characterize them. In this paper we construct the heterotic orbiencoder, a general deep autoencoder to study heterotic orbifold models arising from various Abelian orbifold geometries. Our neural network can be easily trained to successfully encode the large parameter space of many orbifold geometries simultaneously, independently of the statistical dissimilarities of their training features. In particular, we show that our autoencoder is capable of compressing with good accuracy the large parameter space of two promising orbifold geometries in just three parameters. Further, most orbifold models with phenomenologically appealing features appear in bounded regions of this small space. Our contribution hints towards a possible simplification of the classification of (promising) heterotic orbifold models.
format Preprint
id arxiv_https___arxiv_org_abs_2212_00821
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle An autoencoder for heterotic orbifolds with arbitrary geometry
Escalante-Notario, Enrique
Portillo-Castillo, Ignacio
Ramos-Sanchez, Saul
High Energy Physics - Theory
Artificial neural networks have become important to improve the search for admissible string compactifications and characterize them. In this paper we construct the heterotic orbiencoder, a general deep autoencoder to study heterotic orbifold models arising from various Abelian orbifold geometries. Our neural network can be easily trained to successfully encode the large parameter space of many orbifold geometries simultaneously, independently of the statistical dissimilarities of their training features. In particular, we show that our autoencoder is capable of compressing with good accuracy the large parameter space of two promising orbifold geometries in just three parameters. Further, most orbifold models with phenomenologically appealing features appear in bounded regions of this small space. Our contribution hints towards a possible simplification of the classification of (promising) heterotic orbifold models.
title An autoencoder for heterotic orbifolds with arbitrary geometry
topic High Energy Physics - Theory
url https://arxiv.org/abs/2212.00821