SemanticScuttle - klotz.me

klotz: differentiable loss functions*

0 bookmark(s) - Sort by: Date ↓ / Title / - Bookmarks from other users for this tag

1. Enhanced cybersecurity threat detection using novel tri-metaheuristic loss functions in generative adversarial networks with adaptive attention preservation for network traffic augmentation

   This research presents a specialized GAN framework designed to enhance cybersecurity threat detection through advanced network traffic augmentation. By integrating nine differentiable loss components inspired by bio-inspired metaheuristics (Firefly, Jellyfish Search, and Mantis Shrimp), the model resolves class imbalance while preserving critical attack signatures.
   
   * An energy-aware adaptive attention mechanism reduces training energy consumption by 40% without sacrificing accuracy.
   * Tested across seven benchmark datasets, the framework achieved a high average accuracy of 98.73%.
   * The model demonstrated strong robustness against adversarial evasion attempts.

   2026-04-14 Tags: generative adversarial networks, network traffic, cybersecurity, threat detection, differentiable loss functions, adversarial robustness, energy-aware computing, intrusion detection systems, machine learning, feature engineering by klotz

Top of the page

First / Previous / Next / Last / Page 1 of 0