Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

gautierfilardo-efrei/sqnn-adversarial: v.1.1

gautierfilardo-efrei

Published Jun 21, 2026• 0 citations
Publisher

Abstract

Decoherence as Defence and the Magnitude of Noise Regularisation: A Rigorous N -Qubit Theory of Stochastic Quantum Neural Networks for Adversarially Robust Network Intrusion Detection
DOI
10.5281/zenodo.20786869

Related papers

Hybrid Quantum-Classical Neural Networks for Network Intrusion Detection: A Parameter-Efficient Approach

Taha-Amine Miri, Aissa Kerkour Elmiad, Aziz Oukaira +1 more

Lecture notes in networks and systems•Jan 1, 2026•0 citations

Shallow quantum neural networks (SQNNs) with application to crack identification

Meghashrita Das, Arundhuti Naskar, Pabitra Mitra +1 more

Applied Intelligence•Jan 1, 2024•12 citations

Decoherence as Defence and the Magnitude of Noise Regularisation: A Rigorous N -Qubit Theory of Stochastic Quantum Neural Networks for Adversarially Robust Network Intrusion Detection

Gautier-Edouard Filardo

Stochastic quantum neural networks (SQNNs) encode neuronal activations as qubits, synaptic topology as entanglement, and neural noise through a Lindblad master equation. A recent conference study applied a ring-entangled SQNN to collaborative intrusion detection and reached three conclusions: ring entanglement is \emph{essential} for non-local anomaly detection; an adversarial-resilience bound holds but is \emph{conservative}; and the depolarising channel \emph{fails} to act as a dropout-style regulariser, behaving instead as output noise. It left open whether a per-gate stochastic deactivation (``true quantum dropout'') could regularise where the depolarising channel could not, and whether the loose robustness bound could be replaced by a predictive theory. This paper resolves both and extends the framework to real data and to neutral-atom hardware. We give an $N$-qubit formulation through the stochastic master equation and its vectorised Liouvillian, and prove a \emph{decoherence-contraction theorem}: a depolarising channel of strength $γ$ over $L$ entangling layers contracts every weight-$w$ Pauli read-out by a factor $(1-4γ/3)^{wL}$ (for the weight-$1$ read-out used here, $(1-4γ/3)^{L}$); building on the general noise-as-defence result of Du et al., we make this quantitative and operational for intrusion detection. On the real NSL-KDD dataset under white-box FGSM and PGD attacks, a depolarising SQNN trained with the channel is, over seven seeds under strong $\ell_\infty$/$\ell_2$ attacks, significantly more robust than the noiseless circuit ($\ell_\infty$ PGD-$20$, $p=0.04$, large effect) and, critically, never suffers the catastrophic robustness collapse that the noiseless model and gradient-trained classical detectors (which fall from $95\%$ to $47\%$) do, cutting robustness variance roughly twofold; we show this robustness arises from a noise-reshaped training boundary rather than from attack-time gradient contraction. For generalisation, we derive an adaptive-penalty formula showing that per-gate dropout implements a curvature-weighted $L_2$ penalty $\tfrac{p(1-p)}{2}\sumθ^2\partial^2_θL$ in weight space, maximised at $p=1/2$, whereas depolarising noise implements an output-space penalty. A $30$-seed study confirms the formula's quantitative prediction: both mechanisms reduce the train-test gap by a small but statistically significant margin ($\approx\!0.01$; $p<10^{-4}$ and $p=0.004$), are statistically indistinguishable from each other, and the effect is concentrated where overfitting is largest; increasing the dropout rate past $1/2$ does not help, as the formula predicts. The single-seed dichotomy of prior work does not survive replication. We close with a neutral-atom realisation and a feasibility-by-$N$ analysis.

arXiv:2606.24219•Jun 23, 2026•1 citation

Noise, Errors, and Adversarial Robustness in Quantum Machine Learning

Alex Jin, Manas Mukherjee

Quantum science and technology•Jan 1, 2026•1 citation

VQCNN: variational quantum convolutional neural networks based on quantum filters and fully connected layers

Qi Han, Jingtong Wang, Yufan Cui

Abstract Classical machine learning is more susceptible to adversarial examples due to its linear and non-robust nature, which results in a severe degradation of the recognition accuracy of classical machine learning models. Quantum techniques are shown to have a higher robustness advantage and are more resistant to attacks from adversarial examples than classical machine learning. Inspired by the robustness advantage of quantum computing and the feature extraction advantage of convolutional neural networks, this paper proposes a novel variational quantum convolutional neural network model (VQCNN), whose quantum fully connected layer consists of a combination of a quantum filter and a variational quantum neural network to increase the model’s adversarial robustness. The network intrusion detection model based on VQCNN is verified on KDD CUP99 and UNSW-NB datasets. The results show that under the attack of Fast Gradient Sign Method, the decline values of accuracy, precision, and recall rate of the intrusion detection model based on VQCNN are less than those of the other four models, and it has higher adversarial robustness.

The Computer Journal•Jul 6, 2024•10 citations

Adversarially_Robust_QSL_UNSWNB15

Rahul R. Bhoge

A controlled white-box adversarial robustness study of quantum supervised learning for network intrusion detection, demonstrating significant robustness gains from PGD-aware adversarial training while identifying a substantial robustness gap relative to a matched classical neural baseline.

•Aug 25, 2026•0 citations
Stochastic quantum neural networks (SQNNs) encode neuronal activations as qubits, synaptic topology as entanglement, and neural noise through a Lindblad master equation. A recent conference study applied a ring-entangled SQNN to collaborative intrusion detection and reached three conclusions: ring entanglement is \emph{essential} for non-local anomaly detection; an adversarial-resilience bound holds but is \emph{conservative}; and the depolarising channel \emph{fails} to act as a dropout-style regulariser, behaving instead as output noise. It left open whether a per-gate stochastic deactivation (``true quantum dropout'') could regularise where the depolarising channel could not, and whether the loose robustness bound could be replaced by a predictive theory. This paper resolves both and extends the framework to real data and to neutral-atom hardware. We give an NNN-qubit formulation through the stochastic master equation and its vectorised Liouvillian, and prove a \emph{decoherence-contraction theorem}: a depolarising channel of strength γγγ over LLL entangling layers contracts every weight-www Pauli read-out by a factor (1−4γ/3)wL(1-4γ/3)^{wL}(1−4γ/3)wL (for the weight-111 read-out used here, (1−4γ/3)L(1-4γ/3)^{L}(1−4γ/3)L); building on the general noise-as-defence result of Du et al., we make this quantitative and operational for intrusion detection. On the real NSL-KDD dataset under white-box FGSM and PGD attacks, a depolarising SQNN trained with the channel is, over seven seeds under strong ℓ∞\ell_\inftyℓ∞​/ℓ2\ell_2ℓ2​ attacks, significantly more robust than the noiseless circuit (ℓ∞\ell_\inftyℓ∞​ PGD-202020, p=0.04p=0.04p=0.04, large effect) and, critically, never suffers the catastrophic robustness collapse that the noiseless model and gradient-trained classical detectors (which fall from 95%95\%95% to 47%47\%47%) do, cutting robustness variance roughly twofold; we show this robustness arises from a noise-reshaped training boundary rather than from attack-time gradient contraction. For generalisation, we derive an adaptive-penalty formula showing that per-gate dropout implements a curvature-weighted L2L_2L2​ penalty p(1−p)2∑θ2∂θ2L\tfrac{p(1-p)}{2}\sumθ^2\partial^2_θL2p(1−p)​∑θ2∂θ2​L in weight space, maximised at p=1/2p=1/2p=1/2, whereas depolarising noise implements an output-space penalty. A 303030-seed study confirms the formula's quantitative prediction: both mechanisms reduce the train-test gap by a small but statistically significant margin (≈ ⁣0.01\approx\!0.01≈0.01; p<10−4p<10^{-4}p<10−4 and p=0.004p=0.004p=0.004), are statistically indistinguishable from each other, and the effect is concentrated where overfitting is largest; increasing the dropout rate past 1/21/21/2 does not help, as the formula predicts. The single-seed dichotomy of prior work does not survive replication. We close with a neutral-atom realisation and a feasibility-by-NNN analysis.
Abstract Classical machine learning is more susceptible to adversarial examples due to its linear and non-robust nature, which results in a severe degradation of the recognition accuracy of classical machine learning models. Quantum techniques are shown to have a higher robustness advantage and are more resistant to attacks from adversarial examples than classical machine learning. Inspired by the robustness advantage of quantum computing and the feature extraction advantage of convolutional neural networks, this paper proposes a novel variational quantum convolutional neural network model (VQCNN), whose quantum fully connected layer consists of a combination of a quantum filter and a variational quantum neural network to increase the model’s adversarial robustness. The network intrusion detection model based on VQCNN is verified on KDD CUP99 and UNSW-NB datasets. The results show that under the attack of Fast Gradient Sign Method, the decline values of accuracy, precision, and recall rate of the intrusion detection model based on VQCNN are less than those of the other four models, and it has higher adversarial robustness.