Jawdat S. Alkasassbeh

Fast few-shot intrusion detection via depthwise meta-learning

Network intrusion detection systems (NIDS) increasingly rely on deep learning to cope with evolving cyber attacks, yet they typically require large amounts of labeled data per attack type and exhibit poor generalization to novel threats. Recent advances in meta-learning and few-shot learning have shown promise for alle...