Meta-Learning Dynamics Forecasting Using Task Inference
This addresses the generalization issue in dynamics forecasting for applications like fluid dynamics and oceanography, though it appears incremental as it builds on meta-learning with task inference.
The paper tackles the problem of deep learning models struggling to generalize across different domains in dynamics forecasting by proposing DyAd, a model-based meta-learning method that partitions domains into tasks and uses an encoder to infer task features and adapt a forecaster. The method outperforms state-of-the-art approaches on turbulent flow and ocean data forecasting tasks.
Current deep learning models for dynamics forecasting struggle with generalization. They can only forecast in a specific domain and fail when applied to systems with different parameters, external forces, or boundary conditions. We propose a model-based meta-learning method called DyAd which can generalize across heterogeneous domains by partitioning them into different tasks. DyAd has two parts: an encoder which infers the time-invariant hidden features of the task with weak supervision, and a forecaster which learns the shared dynamics of the entire domain. The encoder adapts and controls the forecaster during inference using adaptive instance normalization and adaptive padding. Theoretically, we prove that the generalization error of such procedure is related to the task relatedness in the source domain, as well as the domain differences between source and target. Experimentally, we demonstrate that our model outperforms state-of-the-art approaches on both turbulent flow and real-world ocean data forecasting tasks.