Probing and Enhancing the Robustness of GNN-based QEC Decoders with Reinforcement Learning

2025-08-05

AI Summary

Graph neural network (GNN) decoders for quantum error correction (QEC) exhibit poor robustness against adversarial perturbations, particularly in surface-code decoding. Method: We propose the first reinforcement learning (RL)-based automated vulnerability discovery and adversarial enhancement framework. An RL agent is designed to minimize syndrome bit flips required to induce decoder misclassification, precisely identifying fragile nodes in graph attention network (GAT) decoders; robustness is then improved via end-to-end adversarial training. Contribution/Results: This work introduces the first RL-driven approach for generating adversarial examples and analyzing vulnerabilities of QEC decoders. It further proposes a lightweight, interpretable adversarial training paradigm tailored to surface codes. Experiments on real surface-code data from Google Quantum AI demonstrate that the RL attack achieves high misclassification rates with minimal bit flips, while the adversarially trained decoder reduces logical error rates significantly, substantially improving robustness.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Adversarial Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Graph Neural Networks (GNNs) have emerged as a powerful, data-driven approach for Quantum Error Correction (QEC) decoding, capable of learning complex noise characteristics directly from syndrome data. However, the robustness of these decoders against subtle, adversarial perturbations remains a critical open question. This work introduces a novel framework to systematically probe the vulnerabilities of a GNN decoder using a reinforcement learning (RL) agent. The RL agent is trained as an adversary with the goal of finding minimal syndrome modifications that cause the decoder to misclassify. We apply this framework to a Graph Attention Network (GAT) decoder trained on experimental surface code data from Google Quantum AI. Our results show that the RL agent can successfully identify specific, critical vulnerabilities, achieving a high attack success rate with a minimal number of bit flips. Furthermore, we demonstrate that the decoder's robustness can be significantly enhanced through adversarial training, where the model is retrained on the adversarial examples generated by the RL agent. This iterative process of automated vulnerability discovery and targeted retraining presents a promising methodology for developing more reliable and robust neural network decoders for fault-tolerant quantum computing.

Problem

• Assessing GNN decoder robustness against adversarial perturbations
• Using RL to find minimal syndrome modifications causing misclassification
• Enhancing decoder robustness via adversarial training with RL-generated examples

Innovation

• Reinforcement Learning probes GNN vulnerabilities
• Adversarial training enhances decoder robustness
• Automated vulnerability discovery improves QEC reliability