Non-backtracking Graph Neural Networks

Seonghyun Park,Narae Ryu,Gahee Kim,Dongyeop Woo,Se-Young Yun,Sungsoo Ahn

2024-09-25

Abstract:The celebrated message-passing updates for graph neural networks allow representing large-scale graphs with local and computationally tractable updates. However, the updates suffer from backtracking, i.e., a message flowing through the same edge twice and revisiting the previously visited node. Since the number of message flows increases exponentially with the number of updates, the redundancy in local updates prevents the graph neural network from accurately recognizing a particular message flow relevant for downstream tasks. In this work, we propose to resolve such a redundancy issue via the non-backtracking graph neural network (NBA-GNN) that updates a message without incorporating the message from the previously visited node. We theoretically investigate how NBA-GNN alleviates the over-squashing of GNNs, and establish a connection between NBA-GNN and the impressive performance of non-backtracking updates for stochastic block model recovery. Furthermore, we empirically verify the effectiveness of our NBA-GNN on the long-range graph benchmark and transductive node classification problems.

Machine Learning

What problem does this paper attempt to address?

The paper attempts to address the issue of backtracking in the message-passing update process of Graph Neural Networks (GNNs). Specifically, backtracking occurs when messages pass through the same edge twice and revisit already visited nodes in the graph. This backtracking leads to an exponential increase in the number of message flows with each update, making local updates redundant and affecting the GNN's ability to accurately recognize specific message flows, especially when dealing with large-scale graph data. To mitigate this issue, the authors propose a Non-Backtracking Graph Neural Network (NBA-GNN), which reduces redundancy and improves model performance by prohibiting the reintegration of messages from already visited nodes during the message-passing process.

The main contributions of the paper include:
1. Proposing the NBA-GNN model to address the issue of message flow redundancy in GNNs.
2. Theoretically analyzing how NBA-GNN alleviates the over-squashing phenomenon in GNNs and establishing a connection between NBA-GNN and the superior performance of non-backtracking updates in stochastic block model recovery.
3. Demonstrating the effectiveness of NBA-GNN through experiments on long-range graph benchmarks and node classification tasks, showcasing its state-of-the-art performance on these tasks.