Poster
in
Workshop: AI for Science: Progress and Promises
Continuous PDE Dynamics Forecasting with Implicit Neural Representations
Yuan Yin · Matthieu Kirchmeyer · Jean-Yves Franceschi · Alain Rakotomamonjy · Patrick Gallinari
Keywords: [ generalization ] [ PDEs ] [ Implicit neural representations ] [ continuous models ] [ Dynamical Systems ] [ spatiotemporal forecasting ] [ INRs ] [ physics ] [ partial differential equations ]
Effective data-driven PDE forecasting methods often rely on fixed spatial and / or temporal discretizations. This raises limitations in real-world applications like weather prediction where flexible extrapolation at arbitrary spatiotemporal locations is required. We address this problem by introducing a new data-driven approach, DINo, that models a PDE's flow with continuous-time dynamics of spatially continuous functions. This is achieved by embedding spatial observations independently of their discretization via Implicit Neural Representations in a small latent space temporally driven by a learned ODE. This separate and flexible treatment of time and space makes DINo the first data-driven model to combine the following advantages. It extrapolates at arbitrary spatial and temporal locations; it can learn from sparse irregular grids or manifolds; at test time, it generalizes to new grids or resolutions. DINo outperforms alternative neural PDE forecasters in a variety of challenging generalization scenarios on representative PDE systems.