 |
Center for Nonlinear Studies |
Learning a numeric representation (also known as an embedded vector, or simply embedding)for a piece of binary code (an instruction, a basic block, a function, or even an entire program) has many important security applications, ranging from vulnerability search, plagiarism detection, to malware classification. By reducing a binary code with complex control-flow and data-flow dependencies into a numeric vector using deep learning techniques, we convert complex binarycode detection and search problems into the search of embeddings, which can be done in O(1)time and often can achieve even higher accuracy than traditional methods. In this talk, I amgoing to show how we can revolutionize several security applications using this approach,including vulnerability search, malware variant detection, and binary diffing.
Host: Diego Ray Chavez