Explore and Exploit with Heterotic Line Bundle Models

Abstract

We use deep reinforcement learning to explore a class of heterotic $SU(5)$ GUT models constructed from line bundle sums over Complete Intersection Calabi Yau (CICY) manifolds. We perform several experiments where A3C agents are trained to search for such models. These agents significantly outperform random exploration, in the most favourable settings by a factor of 1700 when it comes to finding unique models. Furthermore, we find evidence that the trained agents also outperform random walkers on new manifolds. We conclude that the agents detect hidden structures in the compactification data, which is partly of general nature. The experiments scale well with $h^{(1,1)}$, and may thus provide the key to model building on CICYs with large $h^{(1,1)}$.

arXiv comment13 pages, 4 figures, python package

References in corpus (26)

• Counting Flux Vacua
• A Mini-Landscape of Exact MSSM Spectra in Heterotic Orbifolds
• An SU(5) Heterotic Standard Model
• A Heterotic Standard Model
• Heterotic Line Bundle Standard Models
• Heterotic GUT and Standard Model Vacua from simply connected Calabi-Yau Manifolds
• Exploring Positive Monad Bundles And A New Heterotic Standard Model
• Heterotic mini-landscape (II): completing the search for MSSM vacua in a Z_6 orbifold
• Computational complexity of the landscape I
• A Comprehensive Scan for Heterotic SU(5) GUT models
• Machine Learning in the String Landscape
• On Free Quotients of Complete Intersection Calabi-Yau Manifolds
• Evolving neural networks with genetic algorithms to study the String Landscape
• Branes with Brains: Exploring String Vacua with Deep Reinforcement Learning
• Machine Learning Line Bundle Cohomologies of Hypersurfaces in Toric Varieties
• Scanning the skeleton of the 4D F-theory landscape
• Machine Learning Line Bundle Cohomology
• Deep learning in the heterotic orbifold landscape
• Counting String Theory Standard Models
• Distinguishing Elliptic Fibrations with AI
• Estimating Calabi-Yau Hypersurface and Triangulation Counts with Equation Learners
• On the computation of non-perturbative effective potentials in the string theory landscape -- IIB/F-theory perspective
• A Zip-code for Quarks, Leptons and Higgs Bosons
• Formulae for Line Bundle Cohomology on Calabi-Yau Threefolds
• Learning non-Higgsable gauge groups in 4D F-theory
• Line bundle cohomologies on CICYs with Picard number two

Cited by in corpus (16)

• Evolving Heterotic Gauge Backgrounds: Genetic Algorithms versus Reinforcement Learning
• Machine Learning Lie Structures & Applications to Physics
• Heterotic String Model Building with Monad Bundles and Reinforcement Learning
• Hilbert Series, Machine Learning, and Applications to Physics
• Deep multi-task mining Calabi-Yau four-folds
• The landscape of promising non-supersymmetric string models
• A Triumvirate of AI Driven Theoretical Discovery
• Ising Machines for Diophantine Problems in Physics
• Generalized Vanishing Theorems for Yukawa Couplings in Heterotic Compactifications
• Characterizing 4-string contact interaction using machine learning
• Heterotic Line Bundle Models on Generalized Complete Intersection Calabi Yau Manifolds
• Universes as Big Data
• Machine Learning on generalized Complete Intersection Calabi-Yau Manifolds
• Hidden Sectors from Multiple Line Bundles for the $B-L$ MSSM
• Deep Learning Calabi-Yau four folds with hybrid and recurrent neural network architectures
• A Database of Calabi-Yau Orientifolds and the Size of D3-Tadpoles