Detecting Overfitting of Machine Learning Techniques for Automatic Vulnerability Detection
Niklas Risse
Abstract
Recent results of machine learning for automatic vulnerability detection have been very promising indeed: Given only the source code of a function f, models trained by machine learning techniques can decide if f contains a security flaw with up to 70% accuracy.
Topics & Concepts
OverfittingComputer scienceVulnerability (computing)Machine learningArtificial intelligenceCode (set theory)Source codeFunction (biology)Artificial neural networkComputer securityOperating systemProgramming languageBiologyEvolutionary biologySet (abstract data type)Advanced Malware Detection TechniquesAdversarial Robustness in Machine LearningNetwork Security and Intrusion Detection