It is difficult to determine the most referenced paper on generative AI, the technology behind OpenAI and GPT-3 (including ChatGPT and Dall-E), as it can vary depending on the specific subfield or application of generative AI being considered. However, one paper that is often cited in the field of generative models is "Generative Adversarial Networks" by Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio, which was published in 2014. This paper introduced the concept of Generative Adversarial Networks (GANs), which has been widely used in various applications such as image synthesis, text generation, and more.
Generative Adversarial Networks
By Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio
(see attached if interested)
Generative_Adversarial_Networks.pdf (1.8 MB)
Abstract
Generative adversarial networks are a kind of artificial intel- ligence algorithm designed to solve the generative model- ing problem. The goal of a generative model is to study a collection of training examples and learn the probability distribution that generated them. Generative Adversarial Networks (GANs) are then able to generate more examples from the estimated probability distribution. Generative models based on deep learning are common, but GANs are among the most successful generative models (espe- cially in terms of their ability to generate realistic high- resolution images). GANs have been successfully applied to a wide variety of tasks (mostly in research settings) but continue to present unique challenges and research opportunities because they are based on game theory while most other approaches to generative modeling are based on optimization.
See Also: