Universal Approximation Property of Invertible Neural Networks

Invertible neural networks (INNs) are neural network architectures with invertibility by design. Thanks to their invertibility and the tractability of their Jacobians, INNs have various machine learning applications such as probabilistic modeling, generative modeling, and representation learning. However, their attractive properties often come at the cost of restricting the layer design, which poses a question on their representation power: can we use these models to approximate sufficiently diverse functions? To answer this question, we have developed a general theoretical framework …

applications approximation architectures cost design generative generative modeling layer machine machine learning machine learning applications modeling network networks neural network neural networks probabilistic modeling property representation representation learning