Cross-Dataset Generalization Results for Adversarially Trained Models

Cross-dataset generalization results present performance metrics for a novel hybrid adversarially-trained deep learning framework. The dataset, authored by Sudip Saha and last updated in June 2026, contains results from experiments on reconnaissance, shellcode, and worms datasets. It reports model accuracy on clean data and under FGSM and PGD adversarial attacks.

Use Cases

• Benchmarking adversarial robustness based on reported accuracy under FGSM and PGD attacks
• Analyzing trade-offs between clean accuracy and robustness based on the described loss components
• Studying model convergence stability based on mentioned training curves and gradient norm decay

Strengths

• Reports specific performance metrics, including 97.88% clean accuracy and 84.9% accuracy under FGSM attacks
• Includes comparative results against CNN and LSTM baselines, showing a 6–10 percentage point improvement
• Licensed under CC-BY-4.0, permitting open sharing and adaptation

Limitations

• Row count is unknown, which may limit suitability assessment
• Column-level documentation is absent; field semantics must be inferred after download
• The dataset is very small at 5.5 KB, indicating limited scope, likely containing only summary results

Provenance

Source

Sudip Saha

Collection Method

Likely contains experimental results from academic research on adversarial machine learning.

Freshness

Last updated 2026-06-01 17:55:13; freshness should be verified