Generative Adversarial Networks are Special Cases of Artificial Curiosity (1990) and also Closely Related to Predictability Minimization (1991)
This work clarifies historical connections between foundational unsupervised learning methods, which is incremental for researchers in machine learning history and theory.
The paper reviews unsupervised neural networks in minimax game settings, showing that Generative Adversarial Networks (GANs) are a specific case of Artificial Curiosity (AC) and are closely related to Predictability Minimization (PM). It corrects a prior claim that PM is not based on a minimax game.
I review unsupervised or self-supervised neural networks playing minimax games in game-theoretic settings: (i) Artificial Curiosity (AC, 1990) is based on two such networks. One network learns to generate a probability distribution over outputs, the other learns to predict effects of the outputs. Each network minimizes the objective function maximized by the other. (ii) Generative Adversarial Networks (GANs, 2010-2014) are an application of AC where the effect of an output is 1 if the output is in a given set, and 0 otherwise. (iii) Predictability Minimization (PM, 1990s) models data distributions through a neural encoder that maximizes the objective function minimized by a neural predictor of the code components. I correct a previously published claim that PM is not based on a minimax game.