Proposes DMD and SINDy as new explainability tools for STGNNs, showing they recover interpretable features like infection times and nodes on semi-synthetic data and action-relevant body parts on real motion data.
arXiv preprint arXiv:2402.07463 , year=
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
representative citing papers
The paper introduces a Common Task Framework for scientific ML, benchmarks it on Kuramoto-Sivashinsky and Lorenz systems, and launches a competition on a global sea surface temperature dataset with holdout data.
MF-SHRED maps point-kinetics trajectories to high-fidelity diffusion fields in one LRA benchmark with ~1-2% errors and ~500x speedup, but two of the three claimed benchmarks are absent.
citing papers explorer
-
Interpreting Temporal Graph Neural Networks with Koopman Theory
Proposes DMD and SINDy as new explainability tools for STGNNs, showing they recover interpretable features like infection times and nodes on semi-synthetic data and action-relevant body parts on real motion data.
-
Common Task Framework For a Critical Evaluation of Scientific Machine Learning Algorithms
The paper introduces a Common Task Framework for scientific ML, benchmarks it on Kuramoto-Sivashinsky and Lorenz systems, and launches a competition on a global sea surface temperature dataset with holdout data.
-
Multi-Fidelity Learning with Shallow Recurrent Decoders for Multi-Physics Applications
MF-SHRED maps point-kinetics trajectories to high-fidelity diffusion fields in one LRA benchmark with ~1-2% errors and ~500x speedup, but two of the three claimed benchmarks are absent.