Deep-Learning-based Vulnerability Detection in Binary Executables

TL;DR

This paper presents a deep learning approach using recurrent neural networks to detect and classify 23 different vulnerabilities in binary executables with high accuracy, expanding beyond previous methods limited to 4 vulnerability types.

Contribution

The study introduces a supervised deep learning method that significantly broadens vulnerability detection capabilities in binary code, achieving high accuracy for multiple vulnerability types.

Findings

Detected 23 vulnerabilities with high accuracy

Achieved 88% accuracy for multi-class classification

Non-vulnerable samples detected with over 98% precision

Abstract

The identification of vulnerabilities is an important element in the software development life cycle to ensure the security of software. While vulnerability identification based on the source code is a well studied field, the identification of vulnerabilities on basis of a binary executable without the corresponding source code is more challenging. Recent research [1] has shown, how such detection can be achieved by deep learning methods. However, that particular approach is limited to the identification of only 4 types of vulnerabilities. Subsequently, we analyze to what extent we could cover the identification of a larger variety of vulnerabilities. Therefore, a supervised deep learning approach using recurrent neural networks for the application of vulnerability detection based on binary executables is used. The underlying basis is a dataset with 50,651 samples of vulnerable code in the form of a standardized LLVM Intermediate Representation. The vectorised features of a Word2Vec model are used to train different variations of three basic architectures of recurrent neural networks (GRU, LSTM, SRNN). A binary classification was established for detecting the presence of an arbitrary vulnerability, and a multi-class model was trained for the identification of the exact vulnerability, which achieved an out-of-sample accuracy of 88% and 77%, respectively. Differences in the detection of different vulnerabilities were also observed, with non-vulnerable samples being detected with a particularly high precision of over 98%. Thus, the methodology presented allows an accurate detection of 23 (compared to 4 [1]) vulnerabilities.