End-to-End Hard-Label Cryptanalytic Model Extraction Using Efficient Sign Recovery
The importance of deep neural networks (DNNs) is widely recognized, and the parameters obtained through training are regarded as valuable assets. Recently, attacks that extract these parameters using only oracle queries to a DNN have been actively studied at IACR conferences. The hard-label setting is the most challeng...
Akira Ito, Takayuki Miura, Yosuke Todo
· 0 citations