Generative Posterior Networks for Approximately Bayesian Epistemic Uncertainty Estimation
This addresses uncertainty estimation for machine learning applications where labeled data is scarce but unlabeled data is abundant, offering a novel method that is not purely incremental.
The paper tackles the problem of estimating epistemic uncertainty in high-dimensional settings with limited labeled data by proposing Generative Posterior Networks (GPNs), which use unlabeled data to approximate Bayesian posteriors, showing empirical improvements in uncertainty estimation and scalability.
In many real-world problems, there is a limited set of training data, but an abundance of unlabeled data. We propose a new method, Generative Posterior Networks (GPNs), that uses unlabeled data to estimate epistemic uncertainty in high-dimensional problems. A GPN is a generative model that, given a prior distribution over functions, approximates the posterior distribution directly by regularizing the network towards samples from the prior. We prove theoretically that our method indeed approximates the Bayesian posterior and show empirically that it improves epistemic uncertainty estimation and scalability over competing methods.