Unlocking Out-of-Distribution Generalization in Dynamics through Physics-Guided Augmentation
This addresses the challenge of out-of-distribution generalization in dynamics modeling for researchers and practitioners, though it appears incremental as it builds on existing augmentation and modeling techniques.
The paper tackles the problem of data scarcity and distribution shifts in dynamical system modeling by proposing SPARK, a physics-guided augmentation plugin that creates physically-plausible training samples, resulting in significant outperformance over state-of-the-art baselines in out-of-distribution and data-scarce scenarios.
In dynamical system modeling, traditional numerical methods are limited by high computational costs, while modern data-driven approaches struggle with data scarcity and distribution shifts. To address these fundamental limitations, we first propose SPARK, a physics-guided quantitative augmentation plugin. Specifically, SPARK utilizes a reconstruction autoencoder to integrate physical parameters into a physics-rich discrete state dictionary. This state dictionary then acts as a structured dictionary of physical states, enabling the creation of new, physically-plausible training samples via principled interpolation in the latent space. Further, for downstream prediction, these augmented representations are seamlessly integrated with a Fourier-enhanced Graph ODE, a combination designed to robustly model the enriched data distribution while capturing long-term temporal dependencies. Extensive experiments on diverse benchmarks demonstrate that SPARK significantly outperforms state-of-the-art baselines, particularly in challenging out-of-distribution scenarios and data-scarce regimes, proving the efficacy of our physics-guided augmentation paradigm.