Physics-aware Spatiotemporal Modules with Auxiliary Tasks for Meta-Learning

Abstract

Modeling the dynamics of real-world physical systems is critical for spatiotemporal prediction tasks, but challenging when data is limited. The scarcity of real-world data and the difficulty in reproducing the data distribution hinder directly applying meta-learning techniques. Although the knowledge of governing partial differential equations (PDE) of data can be helpful for the fast adaptation to few observations, it is mostly infeasible to exactly find the equation for observations in real-world physical systems. In this work, we propose a framework, physics-aware meta-learning with auxiliary tasks, whose spatial modules incorporate PDE-independent knowledge and temporal modules utilize the generalized features from the spatial modules to be adapted to the limited data, respectively. The framework is inspired by a local conservation law expressed mathematically as a continuity equation and does not require the exact form of governing equation to model the spatiotemporal observations. The proposed method mitigates the need for a large number of real-world tasks for meta-learning by leveraging spatial information in simulated data to meta-initialize the spatial modules. We apply the proposed framework to both synthetic and real-world spatiotemporal prediction tasks and demonstrate its superior performance with limited observations.

arXiv commentTo be published in the 30th International Joint Conference on Artificial Intelligence (IJCAI-21)

References in corpus (23)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Inductive Representation Learning on Large Graphs
• MobileNetV2: Inverted Residuals and Linear Bottlenecks
• Matching Networks for One Shot Learning
• A Simple Neural Attentive Meta-Learner
• Neural Ordinary Differential Equations
• Learning data driven discretizations for partial differential equations
• Hamiltonian Neural Networks
• Multimodal Compact Bilinear Pooling for Visual Question Answering and Visual Grounding
• BRITS: Bidirectional Recurrent Imputation for Time Series
• Graph networks as learnable physics engines for inference and control
• Meta-Learning with Latent Embedding Optimization
• Recasting Gradient-Based Meta-Learning as Hierarchical Bayes
• How to train your MAML
• Learned discretizations for passive scalar advection in a 2-D turbulent flow
• Deep Air Quality Forecasting Using Hybrid Deep Learning Framework
• Differentiable Physics-informed Graph Networks
• Modular meta-learning
• DeepSphere: a graph-based spherical CNN
• Joint Modeling of Local and Global Temporal Dynamics for Multivariate Time Series Forecasting with Missing Values
• Coupling Oceanic Observation Systems to Study Mesoscale Ocean Dynamics

Cited by in corpus (1)

• Meta-Learning Dynamics Forecasting Using Task Inference