Motivated by the problem of automated repair of software vulnerabilities, we propose an adversarial learning approach that maps from one discrete source domain to another target domain without requiring paired labeled examples or source and target domains to be bijections. We demonstrate that the proposed adversarial learning approach is an effective technique for repairing software vulnerabilities, performing close to seq2seq approaches that require labeled pairs. The proposed Generative Adversarial Network approach is application-agnostic in that it can be applied to other problems similar to code repair, such as grammar correction or sentiment translation.
@article{arxiv.1805.07475,
title = {Learning to Repair Software Vulnerabilities with Generative Adversarial Networks},
author = {Jacob Harer and Onur Ozdemir and Tomo Lazovich and Christopher P. Reale and Rebecca L. Russell and Louis Y. Kim and Peter Chin},
journal= {arXiv preprint arXiv:1805.07475},
year = {2018}
}
Comments
Presented at 32nd Conference on Neural Information Processing Systems (nips 2018), Montreal Canada