Machine learning for complete intersection Calabi-Yau manifolds: a methodological study

Abstract

We revisit the question of predicting both Hodge numbers $h^{1,1}$ and $h^{2,1}$ of complete intersection Calabi-Yau (CICY) 3-folds using machine learning (ML), considering both the old and new datasets built respectively by Candelas-Dale-Lutken-Schimmrigk / Green-Hübsch-Lutken and by Anderson-Gao-Gray-Lee. In real world applications, implementing a ML system rarely reduces to feed the brute data to the algorithm. Instead, the typical workflow starts with an exploratory data analysis (EDA) which aims at understanding better the input data and finding an optimal representation. It is followed by the design of a validation procedure and a baseline model. Finally, several ML models are compared and combined, often involving neural networks with a topology more complicated than the sequential models typically used in physics. By following this procedure, we improve the accuracy of ML computations for Hodge numbers with respect to the existing literature. First, we obtain 97% (resp. 99%) accuracy for $h^{1,1}$ using a neural network inspired by the Inception model for the old dataset, using only 30% (resp. 70%) of the data for training. For the new one, a simple linear regression leads to almost 100% accuracy with 30% of the data for training. The computation of $h^{2,1}$ is less successful as we manage to reach only 50% accuracy for both datasets, but this is still better than the 16% obtained with a simple neural network (SVM with Gaussian kernel and feature engineering and sequential convolutional network reach at best 36%). This serves as a proof of concept that neural networks can be valuable to study the properties of geometries appearing in string theory.

59 pages; additional details, references updated

References in corpus (29)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Practical Bayesian Optimization of Machine Learning Algorithms
• Going Deeper with Convolutions
• Flux Compactification
• A Model of Inductive Bias Learning
• Markov Chain Monte Carlo and Variational Inference: Bridging the Gap
• Computational complexity of the landscape I
• Machine Learning of Calabi-Yau Volumes
• Machine Learning in the String Landscape
• Evolving neural networks with genetic algorithms to study the String Landscape
• Branes with Brains: Exploring String Vacua with Deep Reinforcement Learning
• Fibrations in CICY Threefolds
• On Algorithmic Universality in F-theory Compactifications
• Topological Invariants and Fibration Structure of Complete Intersection Calabi-Yau Four-Folds
• AdS/CFT as a deep Boltzmann machine
• Genetic Algorithms and the Search for Viable String Vacua
• Deep Learning and AdS/QCD
• Holography as deep learning
• Getting CICY High
• Deep learning black hole metrics from shear viscosity
• Machine Learning $\mathcal{N}=8,D=5$ Gauged Supergravity
• A new N=1 AdS4 Vacuum of Maximal Supergravity
• Inception Neural Network for Complete Intersection Calabi-Yau 3-folds
• Seeing through the String Landscape - a String Hunter's Companion in Particle Physics and Cosmology
• Machine Learning and Algebraic Approaches towards Complete Matter Spectra in 4d F-theory
• Contrast data mining for the MSSM from strings
• Machine Learning Etudes in Conformal Field Theories
• Generalization in multitask deep neural classifiers: a statistical physics approach
• Calabi-Yau Spaces in the String Landscape

Cited by in corpus (17)

• Machine Learning Calabi-Yau Hypersurfaces
• Machine Learning Calabi-Yau Four-folds
• Inception Neural Network for Complete Intersection Calabi-Yau 3-folds
• Applying machine learning to the Calabi-Yau orientifolds with string vacua
• Deep multi-task mining Calabi-Yau four-folds
• A Triumvirate of AI Driven Theoretical Discovery
• Unsupervised Machine Learning Techniques for Exploring Tropical Coamoeba, Brane Tilings and Seiberg Duality
• Machine learning Sasakian and $G_2$ topology on contact Calabi-Yau $7$-manifolds
• Machine Learning Regularization for the Minimum Volume Formula of Toric Calabi-Yau 3-folds
• Characterizing 4-string contact interaction using machine learning
• Deep learning complete intersection Calabi-Yau manifolds
• Universes as Big Data
• Machine Learning on generalized Complete Intersection Calabi-Yau Manifolds
• Constructing and Machine Learning Calabi-Yau Five-folds
• Metastable vacua from torsion and machine learning
• Black holes and the loss landscape in machine learning
• Machine learning of the well known things