Neural Relational Inference with Efficient Message Passing Mechanisms

TL;DR

This paper introduces efficient message passing mechanisms in neural relational inference models, leveraging structural priors and historical data to improve relation prediction and multi-step dynamics in complex systems.

Contribution

It proposes novel message passing techniques and incorporates structural prior knowledge to enhance relation inference and dynamics prediction in neural relational models.

Findings

Outperforms existing methods on simulated physics systems.

Effectively captures coexistence of multiple relations.

Reduces error accumulation in multi-step predictions.

Abstract

Many complex processes can be viewed as dynamical systems of interacting agents. In many cases, only the state sequences of individual agents are observed, while the interacting relations and the dynamical rules are unknown. The neural relational inference (NRI) model adopts graph neural networks that pass messages over a latent graph to jointly learn the relations and the dynamics based on the observed data. However, NRI infers the relations independently and suffers from error accumulation in multi-step prediction at dynamics learning procedure. Besides, relation reconstruction without prior knowledge becomes more difficult in more complex systems. This paper introduces efficient message passing mechanisms to the graph neural networks with structural prior knowledge to address these problems. A relation interaction mechanism is proposed to capture the coexistence of all relations, and a spatio-temporal message passing mechanism is proposed to use historical information to alleviate error accumulation. Additionally, the structural prior knowledge, symmetry as a special case, is introduced for better relation prediction in more complex systems. The experimental results on simulated physics systems show that the proposed method outperforms existing state-of-the-art methods.