SIREN Residual Error as a Regularity Diagnostic for Navier-Stokes Equations
We introduce a method for detecting regularity loss in solutions to the three-dimensional Navier-Stokes equations using the approximation error of Sinusoidal Representation Networks (SIRENs). SIRENs use sin() activations, producing C-infinity outputs that cannot represent non-smooth features. By classical spectral approximation theory, the SIREN error is bounded by O(N^{-s}) where s is the local Sobolev regularity. At a singularity (s to 0), the error is O(1) and localizes via the Gibbs phenomenon. We decompose the velocity field into a cheap analytical baseline (advection-diffusion) and a…
Comments and reviews
No comments yet.
API
curl /api/v1/records/arxiv/2603.18061
curl -X POST /api/v1/records/arxiv/2603.18061/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2603.18061/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"body": "…", "review": true}'