Revealing Decurve Flows for Generalized Graph Propagation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Chen, Ma, Liheng, Chen, Yiyang, Ouyang, Wanli, Bronstein, Michael M., Torr, Philip H. S.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911776311869440
author Lin, Chen
Ma, Liheng
Chen, Yiyang
Ouyang, Wanli
Bronstein, Michael M.
Torr, Philip H. S.
author_facet Lin, Chen
Ma, Liheng
Chen, Yiyang
Ouyang, Wanli
Bronstein, Michael M.
Torr, Philip H. S.
contents This study addresses the limitations of the traditional analysis of message-passing, central to graph learning, by defining {\em \textbf{generalized propagation}} with directed and weighted graphs. The significance manifest in two ways. \textbf{Firstly}, we propose {\em Generalized Propagation Neural Networks} (\textbf{GPNNs}), a framework that unifies most propagation-based graph neural networks. By generating directed-weighted propagation graphs with adjacency function and connectivity function, GPNNs offer enhanced insights into attention mechanisms across various graph models. We delve into the trade-offs within the design space with empirical experiments and emphasize the crucial role of the adjacency function for model expressivity via theoretical analysis. \textbf{Secondly}, we propose the {\em Continuous Unified Ricci Curvature} (\textbf{CURC}), an extension of celebrated {\em Ollivier-Ricci Curvature} for directed and weighted graphs. Theoretically, we demonstrate that CURC possesses continuity, scale invariance, and a lower bound connection with the Dirichlet isoperimetric constant validating bottleneck analysis for GPNNs. We include a preliminary exploration of learned propagation patterns in datasets, a first in the field. We observe an intriguing ``{\em \textbf{decurve flow}}'' - a curvature reduction during training for models with learnable propagation, revealing the evolution of propagation over time and a deeper connection to over-smoothing and bottleneck trade-off.
format Preprint
id arxiv_https___arxiv_org_abs_2402_08480
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Revealing Decurve Flows for Generalized Graph Propagation
Lin, Chen
Ma, Liheng
Chen, Yiyang
Ouyang, Wanli
Bronstein, Michael M.
Torr, Philip H. S.
Machine Learning
Differential Geometry
This study addresses the limitations of the traditional analysis of message-passing, central to graph learning, by defining {\em \textbf{generalized propagation}} with directed and weighted graphs. The significance manifest in two ways. \textbf{Firstly}, we propose {\em Generalized Propagation Neural Networks} (\textbf{GPNNs}), a framework that unifies most propagation-based graph neural networks. By generating directed-weighted propagation graphs with adjacency function and connectivity function, GPNNs offer enhanced insights into attention mechanisms across various graph models. We delve into the trade-offs within the design space with empirical experiments and emphasize the crucial role of the adjacency function for model expressivity via theoretical analysis. \textbf{Secondly}, we propose the {\em Continuous Unified Ricci Curvature} (\textbf{CURC}), an extension of celebrated {\em Ollivier-Ricci Curvature} for directed and weighted graphs. Theoretically, we demonstrate that CURC possesses continuity, scale invariance, and a lower bound connection with the Dirichlet isoperimetric constant validating bottleneck analysis for GPNNs. We include a preliminary exploration of learned propagation patterns in datasets, a first in the field. We observe an intriguing ``{\em \textbf{decurve flow}}'' - a curvature reduction during training for models with learnable propagation, revealing the evolution of propagation over time and a deeper connection to over-smoothing and bottleneck trade-off.
title Revealing Decurve Flows for Generalized Graph Propagation
topic Machine Learning
Differential Geometry
url https://arxiv.org/abs/2402.08480