Probabilistic Programming for Malware Analysis
This work addresses the cyber-defense problem of malware analysis for security professionals, but it appears incremental as it applies a known approach (probabilistic programming) to a specific domain.
The authors tackled the problem of constructing malware lineages by formulating it as a probabilistic model and using a novel probabilistic programming solution to jointly infer lineage and creation times, achieving results that address challenges from large data volumes and obfuscation techniques.
Constructing lineages of malware is an important cyber-defense task. Performing this task is difficult, however, due to the amount of malware data and obfuscation techniques by the authors. In this work, we formulate the lineage task as a probabilistic model, and use a novel probabilistic programming solution to jointly infer the lineage and creation times of families of malware.