Ralts: Robust Aggregation for Enhancing Graph Neural Network Resilience on Bit-flip Errors

TL;DR

This paper introduces Ralts, a lightweight, generalizable method to improve the robustness of graph neural networks against hardware-induced bit-flip errors, significantly enhancing prediction accuracy in safety-critical applications.

Contribution

Ralts is a novel aggregation technique that filters out outliers and recovers graph topology to bolster GNN resilience against bit-flip errors, supporting various models and architectures.

Findings

Ralts improves GNN accuracy by at least 20% under bit-flip errors.

It maintains execution efficiency comparable to standard aggregation functions.

Effective across multiple GNN models, datasets, and error patterns.

Abstract

Graph neural networks (GNNs) have been widely applied in safety-critical applications, such as financial and medical networks, in which compromised predictions may cause catastrophic consequences. While existing research on GNN robustness has primarily focused on software-level threats, hardware-induced faults and errors remain largely underexplored. As hardware systems progress toward advanced technology nodes to meet high-performance and energy efficiency demands, they become increasingly susceptible to transient faults, which can cause bit flips and silent data corruption, a prominent issue observed by major technology companies (e.g., Meta and Google). In response, we first present a comprehensive analysis of GNN robustness against bit-flip errors, aiming to reveal system-level optimization opportunities for future reliable and efficient GNN systems. Second, we propose Ralts, a generalizable and lightweight solution to bolster GNN resilience to bit-flip errors. Specifically, Ralts exploits various graph similarity metrics to filter out outliers and recover compromised graph topology, and incorporates these protective techniques directly into aggregation functions to support any message-passing GNNs. Evaluation results demonstrate that Ralts effectively enhances GNN robustness across a range of GNN models, graph datasets, error patterns, and both dense and sparse architectures. On average, under a BER of $3\times10^{-5}$, these robust aggregation functions improve prediction accuracy by at least 20\% when errors present in model weights or node embeddings, and by at least 10\% when errors occur in adjacency matrices. Ralts is also optimized to deliver execution efficiency comparable to built-in aggregation functions in PyTorch Geometric.