1 paper
Hojjat Aghakhani, Dongyu Meng, Yu-Xiang Wang +2
A recent source of concern for the security of neural networks is the emergence of clean-label dataset poisoning attacks, wherein correctly labeled poison samples are injected into…