Cosmos Week
Map-Level Inference of Primordial Gravitational Waves from the CMB with Spherical Neural Networks
CosmologiaEdição em portuguêsPreprintResultado provisório

Map-Level Inference of Primordial Gravitational Waves from the CMB with Spherical Neural Networks

We train a spherical graph convolutional network to infer the tensor-to-scalar ratio $r$ directly from multifrequency cosmic microwave background $B$-mode maps on the cut sky.

Fonte original citada e enquadrada editorialmente pelo Cosmos Week. arXiv Astrophysics
Assinatura editorialRedação do Cosmos Week
Publicado18 nov 2026 18h59
Atualizado2026-11-18
Tipo de coberturaPreprint
Nível de evidênciaResultado provisório
Leitura4 min de leitura

Pontos-chave

  • Em foco: We train a spherical graph convolutional network to infer the tensor-to-scalar ratio $r$ directly from multifrequency cosmic microwave background
  • Detalhe: Resultado ainda sem revisão por pares
  • Leitura editorial: resultado provisório, ainda sem revisão por pares formal.
Texto completo

We train a spherical graph convolutional network to infer the tensor-to-scalar ratio $r$ directly from multifrequency cosmic microwave background $B$-mode maps on the cut sky. We train a spherical graph convolutional network to infer the tensor-to-scalar ratio $r$ directly from multifrequency cosmic microwave background (CMB) $B$-mode maps on the cut sky.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. ArXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? . Abstract: We train a spherical graph convolutional network to infer the tensor-to-scalar ratio $r$ directly from multifrequency cosmic microwave background (CMB) $B$-mode maps on the cut sky.

We test the network on simulations of a generic ground-based experiment including lensed CMB, inhomogeneous noise, and stochastic Galactic dust and synchrotron emission. We compare the network with a multifrequency bandpower likelihood built from the same 20, 000 simulations used to train it.

Working directly on the maps, with no power-spectrum estimation or component separation, the network reduces the root-mean-square error on $r$ by a factor of 1.3 and gives well-calibrated 68% intervals 1.2 times narrower than the likelihood with the same prior. Saliency maps show that the network learns to favor the deepest regions of the sky and the cleanest frequency channels without being given the survey depth.

Both approaches are tied to the foreground model they are built on: the network generalizes to some alternative foreground models without retraining, but not to a three-dimensional dust model with line-of-sight variations.

Fonte