Graphical RNN Models

TL;DR

This paper presents a flexible RNN-based framework for modeling multiple interacting time series, explicitly capturing entity interactions and incorporating domain knowledge, demonstrated through improved weather prediction results.

Contribution

It introduces a novel, adaptable framework that explicitly models entity interactions in time series and integrates domain knowledge into RNN architectures.

Findings

Achieves better weather prediction accuracy than strong baselines.

Demonstrates the effectiveness of explicit interaction modeling.

Shows how to incorporate domain constraints into RNNs.

Abstract

Many time series are generated by a set of entities that interact with one another over time. This paper introduces a broad, flexible framework to learn from multiple inter-dependent time series generated by such entities. Our framework explicitly models the entities and their interactions through time. It achieves this by building on the capabilities of Recurrent Neural Networks, while also offering several ways to incorporate domain knowledge/constraints into the model architecture. The capabilities of our approach are showcased through an application to weather prediction, which shows gains over strong baselines.