The idea of adversarial training can be dated back to early works in the 1990s, such as Schmidhuber's Learning Factorial Codes by Predictability Minimization (Neural Computation, 1992, 4(6): 863-879). In 2013, adversarial model inferring without any prior information was proposed in A Coevolutionary Approach to Learn Animal Behavior Through Controlled Interaction (Li, et al., Proceedings of the 15th Annual Conference on Genetic and Evolutionary Computation, 2013, 223-230). In 2014, GANs were first introduced by Goodfellow et al. in Generative Adversarial Networks.
Li, et al., the same authors who proposed animal behavior inferring, proposed the term Turing learning in 2016 in Turing learning: a metric-free approach to inferring behavior and its application to swarms (Swarm Intelligence, 10 (3): 211-243). Turing learning is related to the Turing...