November 2019
Intermediate to advanced
346 pages
9h 36m
English
Experts in information security can usually identify potentially exploitable pieces of code. Yet, the work is intensive and costly, and may not be sufficient to make a program secure. One of the great advantages of deep learning over traditional machine learning is that features can be automatically discovered. This allows us to alleviate the need for a human expert on vulnerabilities, as well as to produce more effective systems. In this recipe, we'll utilize a modified version of VulDeePecker : A Deep Learning-Based System for Vulnerability Detection (https://arxiv.org/pdf/1801.01681.pdf), to automatically detect buffer error vulnerabilities and resource ...