Neural network approximations for Calabi-Yau metrics

A bstract Ricci flat metrics for Calabi-Yau threefolds are not known analytically. In this work, we employ techniques from machine learning to deduce numerical K3, the Fermat quintic, and Dwork quintic. This investigation employs a simple, modular neural network architecture that is capable of approximating Kähler manifolds dimensions two three. We show measures assess flatness consistency metric decrease after training. improvement corroborated by performance trained on an independent validation set. Finally, demonstrate learnt showing it invariant under discrete symmetries expected possess.