Generative Adversarial Nets.

Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., et al. (2014). Generative Adversarial Nets. Advances in Neural Information Processing Systems.

Type

Book chapter

Intellectual base

Review or monograph

Year

2014

Pages cited

2672-2680

Link

https://papers.nips.cc/paper/2014/hash/5ca3e9b122f61f8f06494c97b1afccf3-Abstract.html

Cited from

economics_finance, mathematics

Cited by

4 citations across 4 artifacts.

Each citation links to the sentence it supports in the citing article.

Primes

• Minimax Strategy
  • In statistics and machine learning, minimax estimators minimize worst-case risk over the parameter space, robust estimation is minimax over contamination neighborhoods, and adversarial training, certified robustness, and GAN training are explicitly minimax over perturbations or over a generator-discriminator game.

    This sourceFormulates GAN training as an explicit minimax (two-player) game between generator and discriminator; the adversarial-robustness lineage extends to adversarial training.

• Zero Sum Game
  • The pattern ports further. The minimax structure into machine learning: the zero-sum game-theory of optimal play transferred into adversarial training and robust optimisation, where two agents' opposed objectives instantiate the fixed-total structure formally.

    This sourceCasts adversarial training as a two-player minimax game with opposed objectives — the machine-learning instantiation of the fixed-total/minimax structure.

Domain-specific

• Generative adversarial network
  • GAN denotes the adversarial two-model training game rather than every generator, objectives and divergences vary, equilibrium is idealized and training can suffer instability mode collapse and discriminator overfitting.
• Mode collapse
  • A generator can repeatedly map many latent inputs to similar outputs because the training objective rewards locally convincing samples without adequately penalizing omitted regions of the data distribution.

Verification

Does it exist? Not checked yet. This entry carries no identifier to resolve. It was extracted from the citation as written in the article, normalized, and deduplicated against the rest of the registry.

Does it back the claim? Not recorded. Neither this nor any other of the 4 citations of this work carries a recorded support check.

Support is checked per citation rather than per work — the same source can be cited soundly in one article and wrongly in another. Per-citation recording began recently, so a citation with no recorded check is a gap in the record rather than evidence it went unchecked.

See how references were verified.

Links previously used in the corpus

Before the registry existed this work was also linked 1 other way.

• https://papers.nips.cc/paper/5423-generative-adversarial-nets ×1

Registry ID ref:0369bbb88dfe · see in the full table