The generative model is pitted against an adversary
Goodfellow et al., 10 June 2014 (arXiv / NIPS 2014). Ian Goodfellow and coauthors including Yoshua Bengio at Université de Montréal describing the adversarial framework in the introduction of their June 2014 arXiv paper "Generative Adversarial Nets".
More from Goodfellow et al. · AI and science · Before ChatGPT
Who else?
What it refers to: The adversary is a discriminative model that learns whether a sample comes from the model distribution or the data, like police detecting counterfeit currency.
Other voices
- “Machine must fight machine” Nikola Tesla,
- “An imperceptible non-random perturbation to a test image” Szegedy et al.,
- “Human evaluators have difficulty distinguishing from articles written by humans” Brown et al.,
- “World models should not be generative” Yann LeCun, Researcher,
- “So-called generative artificial intelligence is not really generative” Pope Francis, Thinker,
- “The Analytical Engine has no pretensions whatever to originate any thing” Ada Lovelace, Researcher,










