Equivariant Graph Hierarchy-Based Neural Networks

2022/02/22 by Han, Jiaqi, Huang, Wenbing, Xu, Tingyang +1 · 2 citations
#FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2202.10643

Abstract

Equivariant Graph neural Networks (EGNs) are powerful in characterizing the dynamics of multi-body physical systems. Existing EGNs conduct flat message passing, which, yet, is unable to capture the spatial/dynamical hierarchy for complex systems particularly, limiting substructure discovery and global information fusion. In this paper, we propose Equivariant Hierarchy-based Graph Networks (EGHNs) which consist of the three key components: generalized Equivariant Matrix Message Passing (EMMP) , E-Pool and E-UpPool. In particular, EMMP is able to improve the expressivity of conventional equivariant message passing, E-Pool assigns the quantities of the low-level nodes into high-level clusters, while E-UpPool leverages the high-level information to update the dynamics of the low-level nodes. As their names imply, both E-Pool and E-UpPool are guaranteed to be equivariant to meet physic symmetry. Considerable experimental evaluations verify the effectiveness of our EGHN on several applications including multi-object dynamics simulation, motion capture, and protein dynamics modeling.

Cited by

• Learning Equivariant Functions via Quadratic Forms
• Fast and Distributed Equivariant Graph Neural Networks by Virtual Node Learning

Related

• Fast and Distributed Equivariant Graph Neural Networks by Virtual Node Learning
• E(n) Equivariant Graph Neural Networks
• Geometric and Physical Quantities Improve E(3) Equivariant Message Passing
• Are High-Degree Representations Really Unnecessary in Equivariant Graph Neural Networks?
• Equivariant Spatio-Temporal Attentive Graph Networks to Simulate Physical Dynamics