Adaptive Initial Residual Connections for GNNs with Theoretical Guarantees

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Shirzadi, Mohammad, Dehkordi, Ali Safarpoor, Zehmakan, Ahad N.
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911257248923648
author Shirzadi, Mohammad
Dehkordi, Ali Safarpoor
Zehmakan, Ahad N.
author_facet Shirzadi, Mohammad
Dehkordi, Ali Safarpoor
Zehmakan, Ahad N.
contents Message passing is the core operation in graph neural networks, where each node updates its embeddings by aggregating information from its neighbors. However, in deep architectures, this process often leads to diminished expressiveness. A popular solution is to use residual connections, where the input from the current (or initial) layer is added to aggregated neighbor information to preserve embeddings across layers. Following a recent line of research, we investigate an adaptive residual scheme in which different nodes have varying residual strengths. We prove that this approach prevents oversmoothing; particularly, we show that the Dirichlet energy of the embeddings remains bounded away from zero. This is the first theoretical guarantee not only for the adaptive setting, but also for static residual connections (where residual strengths are shared across nodes) with activation functions. Furthermore, extensive experiments show that this adaptive approach outperforms standard and state-of-the-art message passing mechanisms, especially on heterophilic graphs. To improve the time complexity of our approach, we introduce a variant in which residual strengths are not learned but instead set heuristically, a choice that performs as well as the learnable version.
format Preprint
id arxiv_https___arxiv_org_abs_2511_06598
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adaptive Initial Residual Connections for GNNs with Theoretical Guarantees
Shirzadi, Mohammad
Dehkordi, Ali Safarpoor
Zehmakan, Ahad N.
Machine Learning
Message passing is the core operation in graph neural networks, where each node updates its embeddings by aggregating information from its neighbors. However, in deep architectures, this process often leads to diminished expressiveness. A popular solution is to use residual connections, where the input from the current (or initial) layer is added to aggregated neighbor information to preserve embeddings across layers. Following a recent line of research, we investigate an adaptive residual scheme in which different nodes have varying residual strengths. We prove that this approach prevents oversmoothing; particularly, we show that the Dirichlet energy of the embeddings remains bounded away from zero. This is the first theoretical guarantee not only for the adaptive setting, but also for static residual connections (where residual strengths are shared across nodes) with activation functions. Furthermore, extensive experiments show that this adaptive approach outperforms standard and state-of-the-art message passing mechanisms, especially on heterophilic graphs. To improve the time complexity of our approach, we introduce a variant in which residual strengths are not learned but instead set heuristically, a choice that performs as well as the learnable version.
title Adaptive Initial Residual Connections for GNNs with Theoretical Guarantees
topic Machine Learning
url https://arxiv.org/abs/2511.06598