Machine Learning Calabi-Yau Three-Folds, Four-Folds, and Five-Folds

Abstract

In this manuscript, we demonstrate, using several regression techniques, that the remaining independent Hodge numbers of complete intersection Calabi-Yau four-folds and five-folds can be machine learned from $h^{1,1}$ and $h^{2,1}$. Consequently, we combine the Hodge numbers $h^{1,1}$ and $h^{2,1}$ from the complete intersection Calabi-Yau three-folds, four-folds, and five-folds into a single dataset. We then implement various classification algorithms on this dataset. For example, Gaussian process and naive Bayes classifiers both achieve $100\%$ accuracy in binary classification between three-folds and four-folds. Using the Support Vector Machine (SVM) algorithm, a special corner is identified in the Calabi-Yau four-fold landscape (characterized by $15\le h^{1,1}\le 30$ and $95\le h^{2,1}\le 100$) during multiclass classification. Furthermore, the highest accuracy $1.00000$, in classifying Calabi-Yau three-folds, four-folds, and five-folds is obtained using the naive Bayes classifier.

arXiv comment10 pages, 11 figures

References in corpus (12)

• CFT's From Calabi-Yau Four-folds
• Flux compactifications in string theory: a comprehensive review
• Machine Learning CICY Threefolds
• All Complete Intersection Calabi-Yau Four-Folds
• Topological Invariants and Fibration Structure of Complete Intersection Calabi-Yau Four-Folds
• Machine learning Calabi-Yau metrics
• Getting CICY High
• Machine Learning Calabi-Yau Hypersurfaces
• Machine Learning Calabi-Yau Four-folds
• M-theory on Calabi-Yau Five-Folds
• A Mirror Pair of Calabi-Yau Fourfolds in Type II String Theory
• Machine learning complete intersection Calabi-Yau 3-folds