Generative adversarial networks (GANs) — inspires but does not transfer cleanly to → Adversarial reward functions

Adversarial reward functions is explicitly modeled on GANs: instead of a static reward function sitting still to be gamed, a second agent actively searches for scenarios the primary system rates highly but a human would rate poorly, the same generator-versus-discriminator shape Goodfellow et al. (2014) introduced (concrete-problems, §"Adversarial Reward Functions:", p. 9). The analogy has a limit the paper is careful to flag. In a GAN, generator and discriminator are trained symmetrically from the same gradient signal and can be arbitrarily matched in capability; the setup is stable partly because neither side needs a built-in advantage. The reward-checking agent in adversarial reward functions cannot be symmetric in that way: the paper notes it "would have to ensure that the reward-checking agent is more powerful, in a somewhat subtle sense, than the agent that is trying to achieve rewards" (concrete-problems, §"Adversarial Reward Functions:", p. 9), an asymmetric power requirement with no GAN analogue and no proposed mechanism for guaranteeing it holds as the reward-seeking agent scales up.