Characterizing 4-string contact interaction using machine learning

Abstract

The geometry of 4-string contact interaction of closed string field theory is characterized using machine learning. We obtain Strebel quadratic differentials on 4-punctured spheres as a neural network by performing unsupervised learning with a custom-built loss function. This allows us to solve for local coordinates and compute their associated mapping radii numerically. We also train a neural network distinguishing vertex from Feynman region. As a check, 4-tachyon contact term in the tachyon potential is computed and a good agreement with the results in the literature is observed. We argue that our algorithm is manifestly independent of number of punctures and scaling it to characterize the geometry of $n$-string contact interaction is feasible.

28+10 pages, 13 figures, 6 tables

References in corpus (15)

• Evolving neural networks with genetic algorithms to study the String Landscape
• Branes with Brains: Exploring String Vacua with Deep Reinforcement Learning
• Neural Networks and Quantum Field Theory
• String Field Theory -- A Modern Introduction
• Precision Machine Learning
• From Symmetric Product CFTs to ${\mathrm{AdS}}_3$
• Hyperbolic three-string vertex
• Nonperturbative renormalization for the neural network-QFT correspondence
• Closed Bosonic String Field Theory at Quintic Order II: Marginal Deformations and Effective Potential
• Moduli-dependent KK towers and the swampland distance conjecture on the quintic Calabi-Yau manifold
• Deep multi-task mining Calabi-Yau four-folds
• Calabi-Yau Metrics, Energy Functionals and Machine-Learning
• Twistor Coverings and Feynman Diagrams
• Connections between reflected entropies and hyperbolic string vertices
• Classical geometry from the tensionless string

Cited by in corpus (4)

• Open string stub as an auxiliary string field
• Boundary terms in string field theory
• Topological recursion for hyperbolic string field theory
• String Scattering and Evolution of Ryu-Takayanagi Surface