Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

An integrated evolution-aware meta-learning framework with adversarial morphological augmentation for zero-day threat

Kavitha Lanka1, Kareemulla Shaik1

  • 1School of Computer Science Engineering, VIT-A.P University, Amaravati, India.

Frontiers in Artificial Intelligence
|June 18, 2026
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Introduction to Learning01:18

Introduction to Learning

Learning is the process of acquiring knowledge or skills through practice or experience, leading to long-lasting behavioral changes. This acquisition occurs through interaction with the environment and requires practice or experience. For instance, mastering a skill such as surfing requires considerable practice and experience, highlighting the essential role of repeated interactions with the environment in learning.
In contrast to learned behaviors, unlearned behaviors such as crying, sexual...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Optimized feature selection and zero-parameter channel attention BiLSTM for RPL-attack classification in IoT networks.

PloS one·2026
See all related articles

M³-GAZE enhances zero-day threat detection by treating it as an evolutionary problem. This novel framework improves adaptability and reduces false negatives for evolving cyber threats.

Area of Science:

  • Cybersecurity and Artificial Intelligence
  • Machine Learning for Anomaly Detection

Background:

  • Traditional threat detection methods struggle with polymorphic, adaptive, and zero-day attacks due to reliance on static signatures and historical data.
  • Expanding attack surfaces in cloud and IoT environments necessitate advanced detection beyond static similarity matching.
  • Existing deep learning and anomaly detection approaches face limitations such as data scarcity, concept drift, poor interpretability, and weak generalization.

Purpose of the Study:

  • To introduce M³-GAZE (Meta-Morphological GAN-Augmented Zero-day Detection Engine), a novel framework for evolutionary threat detection.
  • To address the limitations of current threat detection systems in identifying and adapting to novel and evolving cyber threats.
  • To enable accurate zero-day detection with limited data and provide interpretable, temporally aware threat analysis.
Keywords:
adversarial data augmentationanalysisevolutionary threat modelingmeta-learninguncertainty-aware detectionzero-day attacks

Related Experiment Videos

Main Methods:

  • Latent Morphology Spectrum Extraction (LMSE) to learn continuous latent representations of threat structural invariants.
  • Adversarial Evolutionary GAN (AE-GAN) to generate evolution-consistent synthetic samples for unseen attack variations.
  • Meta-Adaptation Framework for Threat Intelligence (MAFTI) for learning adaptation strategies across evolutionary tasks.
  • Adversarial Uncertainty Calibration Layer (AUCL) to evaluate epistemic uncertainty via adversarial perturbations.
  • Causal Evolutionary Threat Graph Synthesizer (CETGS) to construct temporal causal graphs for threat evolution explanation.

Main Results:

  • M³-GAZE demonstrates improved zero-day recall and enhanced few-shot adaptability.
  • The framework effectively calibrates uncertainty and reduces false negatives in threat detection.
  • Provides interpretable and temporally aware insights into threat evolution and propagation dynamics.

Conclusions:

  • M³-GAZE offers a robust solution for detecting evolving and zero-day cyber threats.
  • The evolutionary inference approach significantly enhances the adaptability and accuracy of threat detection systems.
  • The framework's ability to provide interpretable, uncertainty-calibrated, and temporally aware detection marks a significant advancement in cybersecurity.