Continuous PDE Dynamics Forecasting with Implicit Neural Representations

Authors: Yuan Yin, Matthieu Kirchmeyer, Jean-Yves Franceschi, Alain Rakotomamonjy, patrick gallinari

ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate DINO s versatility and state-of-the-art performance versus prior neural PDE forecasters, representative of grid, operator and INR-based methods, via thorough experiments on challenging multi-dimensional PDEs in various spatiotemporal generalization settings.
Researcher Affiliation Collaboration 1Sorbonne Université, CNRS, ISIR, F-75005 Paris, France 2Criteo AI Lab, Paris
Pseudocode Yes Algorithm 1: DINO pseudo-code
Open Source Code Yes We provide our source code at https://github.com/ mkirchmeyer/DINo.
Open Datasets Yes We evaluate DINO on real-world data to further assess its applicability. Following de Bézenac et al. (2018) and Donà et al. (2021), we model the Sea Surface Temperature (SST) of the Atlantic ocean, derived from the data-assimilation engine NEMO (Nucleus for European Modeling of the Ocean, Madec & NEMO System Team) using E.U. Copernicus Marine Service Information.1 https://data.marine.copernicus.eu/product/GLOBAL_ANALYSIS_FORECAST_ PHY_001_024/description.
Dataset Splits No The paper specifies train and test splits (e.g., '512 train trajectories and 32 test trajectories'), but does not explicitly mention or define a separate validation set or split percentages for reproduction.
Hardware Specification Yes This represents 2.76 s for 64 trajectories on a single Tesla V100 Nvidia GPU.
Software Dependencies No The paper mentions 'Py Torch (Paszke et al., 2019)' and 'Torch Diff Eq (Chen et al., 2018)', but does not provide specific version numbers for these software components.
Experiment Setup Yes We list the hyperparameters of DINO for each dataset in Table 7.