Poster
in
Affinity Workshop: LatinX in AI (LXAI) Research Workshop
A Self-Interpretable Obfuscated Malware Detection System based on Memory Analysis and Self-Supervised Tabular Learning
Josue Genaro Almaraz-Rivera · Jose Antonio Cantoral-Ceballos · Juan Botero · Jesus Perez-Diaz
Keywords: [ Tabular Learning ] [ Obfuscated Malware ] [ Tabular Networks ] [ Memory Analysis ] [ Explainable Artificial Intelligence ] [ self-supervised learning ]
Obfuscated malware detection is a complex task where classification performance is seriously affected due to the evasion techniques presented in the input software samples. This research follows a novel memory analysis technique to examine features extracted from different RAM snapshots over a compromised Windows Virtual Machine. The Self-Supervised Learning paradigm is selected as a novel training strategy for the representation learning of massive amounts of unlabeled information with strong model adaptation capabilities to unseen data. To the best of our knowledge, this is the first work implementing Self-Supervised Learning directly in the tabular data domain for the malware detection problem.