Daniel Gibert, PhD

Daniel Gibert, PhD

Researcher at the Artificial Intelligence Research Institute (IIIA-CSIC)

HYDRA: A multimodal deep learning framework for malware classification

Published in Journal Computers & Security, 2020

This paper proposes a multimodal deep learning system to categorize malware into families that involves multiple modalities of data:

  • The list of Windows API functions calls.
  • The sequence of assembly language instructions representing malware’s assembly language source code.
  • The sequence of hexadecimal values representing malware’s binary content.

HYDRA.jpg

HYDRA_results.png

Recommended citation: Daniel Gibert, Carles Mateu, Jordi Planes. (2020). "HYDRA: A multimodal deep learning framework for malware classification." Journal Computers & Security.
Download Paper