PoiCGAN: A Targeted Poisoning Based on Feature-Label Joint Perturbation in Federated Learning

📰 ArXiv cs.AI

PoiCGAN is a targeted poisoning attack method for federated learning that uses feature-label joint perturbation

advanced Published 26 Mar 2026
Action Steps
  1. Understand the concept of federated learning and its vulnerability to poisoning attacks
  2. Analyze the existing poisoning attack methods and their limitations
  3. Implement PoiCGAN to generate targeted poisoning attacks using feature-label joint perturbation
  4. Evaluate the effectiveness of PoiCGAN in compromising federated learning models
Who Needs to Know This

Machine learning engineers and researchers on a team working with federated learning models can benefit from understanding PoiCGAN to improve model security and robustness

Key Insight

💡 PoiCGAN uses feature-label joint perturbation to bypass existing defenses and compromise federated learning models

Share This
💡 PoiCGAN: A new targeted poisoning attack for federated learning #AI #ML
Read full paper → ← Back to News