Predictive Complexity Priors

Eric Nalisnick · Jonathan Gordon · Jose Miguel Hernandez-Lobato

Keywords: [ Applications ] [ Video Analysis ] [ Learning Theory and Statistics ] [ Bayesian Methods ]

[ Abstract ]
Wed 14 Apr 6 a.m. PDT — 8 a.m. PDT


Specifying a Bayesian prior is notoriously difficult for complex models such as neural networks. Reasoning about parameters is made challenging by the high-dimensionality and over-parameterization of the space. Priors that seem benign and uninformative can have unintuitive and detrimental effects on a model's predictions. For this reason, we propose predictive complexity priors: a functional prior that is defined by comparing the model's predictions to those of a reference model. Although originally defined on the model outputs, we transfer the prior to the model parameters via a change of variables. The traditional Bayesian workflow can then proceed as usual. We apply our predictive complexity prior to high-dimensional regression, reasoning over neural network depth, and sharing of statistical strength for few-shot learning.

Chat is not available.