SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2111.06841 · NeurIPS · 2021

A posteriori learning of quasi-geostrophic turbulence parametrization: an experiment on integration steps

Hugo Frezat, Julien Le Sommer, Ronan Fablet, Guillaume Balarac

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 3 functions out of this paper's own repositories and ran 3 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
hrkz/torchqg canonical 3 of 3
FunctionStatusWhere it lives
aliased_wavenumbers Ran hrkz/torchqg/src/grid.py
code served (permissive licence) · get_code("e001266f99fdb820")
to_physical Ran hrkz/torchqg/qg.py
code served (permissive licence) · get_code("e400264f97dbcecf")
to_spectral Ran hrkz/torchqg/qg.py
code served (permissive licence) · get_code("fc72f7a04ec07251")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Modeling the subgrid-scale dynamics of reduced models is a long standing open problem that finds application in ocean, atmosphere and climate predictions where direct numerical simulation (DNS) is impossible. While neural networks (NNs) have already been applied to a range of three-dimensional flows with success, two dimensional flows are more challenging because of the backscatter of energy from small to large scales. We show that learning a model jointly with the dynamical solver and a meaningful a posteriori-based loss function lead to stable and realistic simulations when applied to quasi-geostrophic turbulence. Recently, neural networks (NNs) have been proposed as a promising alternative to algebraic parametrizations in three-dimensional incompressible turbulence [6,15,5]. Two-dimensional problems, however, are more challenging due to the inverse cascade of energy that leads to negative viscosities, and it has been demonstrated that numerical stability of the trained model in decaying turbulence requires either the removal of negative eddy viscosities [13] or a large training dataset [9].

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2111.06841")
get_code_for_paper("2111.06841")
have("2111.06841")

Connect an agent — have() is free.