Gradient Visualization for General Characterization in Profiling Attacks
Gradient Visualization for General Characterization in Profiling Attacks
Loïc Masure,Cécile Canovas,E. Prouff
2019 · DOI: 10.1007/978-3-030-16350-1_9
IACR Cryptology ePrint Archive · 66 Citations
TLDR
A new tool called Gradient Visualization is introduced that aims to proceed a post-mortem information leakage characterization after the successful training of a neural network, which relies on the computation of the gradient of the loss function used during the training.
