← Papers
Graphs

GraphCast

Learning skillful medium-range global weather forecasting

Paper title GraphCast: Learning skillful medium-range global weather forecasting

Lam et al. (Google DeepMind) · Weather & climate · Public: code, weights

Paper arXiv 2212.12794PDF

An encode–process–decode graph network. The encoder maps the latitude–longitude grid onto a multi-mesh, built by refining an icosahedron six times and keeping the edges of every level; 16 message-passing layers run there, and the decoder maps back as a residual update. Each step is six hours, rolled out in sequence.

Published
Science

Share on LinkedIn, X or Bluesky, or or .

Know related work that belongs here?

Something wrong on this page?

More papers

All papers
  1. GraphsR-BiLoGo — Transfer learning for rib effects in sheet-forming surrogates
  2. GraphsRead-Write-Relax — Why neural PDE surrogates need both global and local processing
  3. GraphsDGN / LDGN — Sampling flow distributions on meshes with diffusion graph networks