Adversarial attacks on graph-level embedding methods: a case study

• Published in: Annals of mathematics and artificial intelligence Vol. 91; no. 2-3; pp. 259 - 285
• Main Authors: Giordano, Maurizio, Maddalena, Lucia, Manzo, Mario, Guarracino, Mario Rosario
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.06.2023; Springer; Springer Nature B.V
• Subjects: Algorithms; Analysis; Artificial Intelligence; Case studies; Classification; Complex Systems; Computational linguistics; Computer Science; Cybersecurity; Embedding; Graph representations; Language processing; Machine learning; Mathematics; Methods; Natural language interfaces; Neural networks; Training; 68T01; Adversarial machine learning; Graph embedding; Graph neural networks; 68T07; Graph classification; 68R10; 92B20; Adversarial attacks
• ISSN: 1012-2443; 1573-7470
• DOI: 10.1007/s10472-022-09811-4

As the number of graph-level embedding techniques increases at an unprecedented speed, questions arise about their behavior and performance when training data undergo perturbations. This is the case when an external entity maliciously alters training data to invalidate the embedding. This paper expl...