With machine learning, programs “learn” how to perform complex analytical tasks without direct control by a human. This will lead to the automation of tasks that involve classification, prediction, and anomaly detection. It has great potential to increase security experts’ productivity and efficiency in security operations, program analysis, and many other areas.
Why we should work together
We can help your organization to take advantage of new, fundamental and proprietary research on the application of machine learning to computer security problems. For example:
Apply algorithms developed for natural language processing to the analysis of assembly languages and code. Overcome current barriers in program analysis, such as state-space explosion, to observe program behavior at a higher level than is possible today, and to determine the intent of the code’s creator with greater clarity. Improve the capacity of software to handle more of your security teams’ tasks, so that they can then focus on deeper problems.
This is just the start. As we integrate machine learning into several of our tools – Manticore, Slither, Echidna, and Insight – we will continue to expand our capacity to help you, and the field of research.
What we’ve contributed
- Toward large-scale vulnerability discovery using Machine Learning introduces VDiscover, an open-source tool that performs a fully automatic predictive approach to identify likely exploitable test cases without requiring source code.
- Exniffer: Learning to Prioritize Crashes by Assessing the Exploitability from Memory Dump presents a technique to identify security-critical crashes by applying Machine Learning on a set of features derived from core-dump files and runtime information obtained from hardware-assisted monitoring such as the Last Branch Record register.
Contact us with your machine learning needs.