Conformal Fields from Neural Networks

Abstract

We use the embedding formalism to construct conformal fields in $D$ dimensions, by restricting Lorentz-invariant ensembles of homogeneous neural networks in $(D+2)$ dimensions to the projective null cone. Conformal correlators may be computed using the parameter space description of the neural network. Exact four-point correlators are computed in a number of examples, and we perform a 4D conformal block decomposition that elucidates the spectrum. In some examples the analysis is facilitated by recent approaches to Feynman integrals. Generalized free CFTs are constructed using the infinite-width Gaussian process limit of the neural network, enabling a realization of the free boson. The extension to deep networks constructs conformal fields at each subsequent layer, with recursion relations relating their conformal dimensions and four-point functions. Numerical approaches are discussed.

v1: 48 pages; v2 60 pages (journal version)

References in corpus (6)

• Solving the 3d Ising Model with the Conformal Bootstrap II. c-Minimization and Precise Critical Exponents
• Lectures on differential equations for Feynman integrals
• Bootstrapping Mixed Correlators in the 3D Ising Model
• Bounds in 4D Conformal Field Theories with Global Symmetry
• Five dimensional $O(N)$-symmetric CFTs from conformal bootstrap
• Bootstrapping the 3d Ising Stress Tensor

Cited by in corpus (6)

• Holographic flavour and neural networks
• Gravity Dual of Networks
• On the solution of Euclidean path integrals with neural networks
• Gauge-covariant stochastic neural fields: Stability and finite-width effects
• Approximating Grassmann valued path integrals with radial basis function neural networks
• Conformal Defects in Neural Network Field Theories