Detecting Overfitting of Machine Learning Techniques for Automatic Vulnerability Detection

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.

BibTeX

@inproceedings{Risse:FSE23,
  author    = {Niklas Risse},
  title     = {Detecting Overfitting of Machine Learning Techniques for Automatic Vulnerability Detection},
  booktitle = {{ESEC/SIGSOFT} {FSE}},
  pages     = {2189--2191},
  publisher = {{ACM}},
  year      = {2023},
}

Detecting Overfitting of Machine Learning Techniques for Automatic Vulnerability Detection

Abstract

Related papers