<ul><li>Link prediction in graph data utilizes various algorithms and machine learning/deep learning models to predict potential relationships between graph nodes.</li><li>Recent research highlights the vulnerability of link prediction models to adversarial attacks, such as poisoning and evasion attacks.</li><li>A new approach using meta-learning techniques is proposed to exploit Variational Graph Auto-Encoder (VGAE) model's link prediction performance.</li><li>Comprehensive experiments demonstrate that the proposed approach significantly diminishes link prediction performance and outperforms other state-of-the-art methods.</li></ul>

Exploiting Meta-Learning-based Poisoning Attacks for Graph Link Prediction

Discover more