Study Reveals CNN Architectures More Robust Than Random Forest in Adversarial Network Intrusion Detection
A new preprint study tested three machine learning architectures used in Network Intrusion Detection Systems against adversarial attacks, finding that a Random Forest model with near-perfect baseline accuracy dropped 73 percentage points in performance under even minimal attack pressure. The research used the ACI-IoT-2023 dataset with over 1.2 million samples and applied two established adversarial attack methods across a range of perturbation strengths. The findings challenge the assumption that high baseline accuracy is a reliable indicator of real-world robustness in security-critical deployments.
Researchers from arXiv's cs.CR/cs.LG community evaluated three popular machine learning classifier architectures — a 1D Convolutional Neural Network (CNN), a Long Short-Term Memory (LSTM) network, and a Random Forest (RF) ensemble — for their resilience against adversarial attacks in the context of Network Intrusion Detection Systems (NIDS). Using the ACI-IoT-2023 dataset spanning over 1.2 million samples across 12 attack categories, each model was subjected to Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) attacks at perturbation budgets ranging from ε=0.01 to ε=0.1. The Random Forest model achieved the highest baseline accuracy at 99.98%, but suffered a catastrophic 73 percentage-point drop at the smallest perturbation tested, while the CNN retained 95.5% accuracy at ε=0.01 and degraded more gradually as perturbation increased. LSTM performance fell between the two extremes. The authors argue these results overturn conventional wisdom that equates high baseline accuracy with deployment readiness, and recommend CNN-based architectures for adversarial environments, along with scenario-specific deployment guidance.
What's missing
The adversarial attacks are applied in normalized feature space, and it remains an open question whether such perturbations are fully realizable in live network traffic without violating protocol constraints. The study does not evaluate adversarial training or other defensive techniques as mitigations, nor does it test architectures beyond the three selected. Generalizability to datasets other than ACI-IoT-2023 is unconfirmed.
What different sources said
- arXiv cs.LGCenter
Categorical Robustness Assessment for Machine Learning based Network Intrusion Detection Systems
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.