AInstein: Numerical Einstein Metrics via Machine Learning

Abstract

A new semi-supervised machine learning package is introduced which successfully solves the Euclidean vacuum Einstein equations with a cosmological constant, without any symmetry assumptions. The model architecture contains subnetworks for each patch in the manifold-defining atlas. Each subnetwork predicts the components of a metric in its associated patch, with the relevant Einstein conditions of the form $R_{\mu \nu }-\lambda g_{\mu \nu }=0$ being used as independent loss components (here $\mu ,\nu =1,2,\cdots ,n$, where $n$ is the dimension of the Riemannian manifold, and the Einstein constant $\lambda \in \{+1,0,-1\}$). To ensure the consistency of the global structure of the manifold, another loss component is introduced across the patch subnetworks which enforces the coordinate transformation between the patches, $g^{\prime }=J^{T}gJ$, for an appropriate analytically known Jacobian $J$. We test our method for the case of spheres represented by a pair of patches in dimensions 2, 3, 4, and 5. In dimensions 2 and 3, the geometries have been fully classified. However, it is unknown whether a Ricci-flat metric can exist on spheres in dimensions 4 and 5. This work hints against the existence of such a metric.

24 pages; 11 figures; 5 tables. v3. Added journal accepted version. v2. Correct typos, make notation consistent for losses, update abstract and conclusion

References in corpus (14)

• Evolution of Binary Black Hole Spacetimes
• Static Axisymmetric Vacuum Solutions and Non-Uniform Black Strings
• Numerical solution of gravitational dynamics in asymptotically anti-de Sitter spacetimes
• Black Strings, Low Viscosity Fluids, and Violation of Cosmic Censorship
• A new approach to static numerical relativity, and its application to Kaluza-Klein black holes
• Numerical Methods for Finding Stationary Gravitational Solutions
• Ricci solitons, Ricci flow, and strongly coupled CFT in the Schwarzschild Unruh or Boulware vacua
• GRChombo : Numerical Relativity with Adaptive Mesh Refinement
• Machine learning Calabi-Yau metrics
• Six-dimensional localized black holes: numerical solutions
• CYJAX: A package for Calabi-Yau metrics with JAX
• Bayesian Renormalization
• Machine Learned Calabi-Yau Metrics and Curvature
• Exotic Spheres' Metrics and Solutions via Kaluza-Klein Techniques