Adversarial Attacks are Reversible with Natural Supervision.

Chengzhi Mao Mia Chiquier Hao Wang Junfeng Yang Carl Vondrick

Published in: ICCV (2021)

Keyphrases

markov chain
active learning
cellular automata
real world
learning algorithm
countermeasures
database
search engine
multiresolution
security protocols
attack detection
chosen plaintext