Thesis topic:

Malicious PyTorch Models: A Digital Forensics Perspective

• Supervisor: Lukas Daubner
  • contact: lukas.daubner@ut.ee
• The aim of the thesis is to examine the attacks utilising malicious PyTorch models as a vector. Then, to identify the forensic artefacts tied to both the malicious models and models in general. Finally, the knowledge shall be utilised to develop a prototype tool to extract the artefacts from the model.

<< back