Variational Learning of Fractional Posteriors
Provides a new variational framework for probabilistic modeling that improves posterior calibration and generation quality, particularly for VAEs.
This work introduces a variational objective for learning fractional posteriors, demonstrating improved calibration in mixture models and higher evidence bounds in variational autoencoders (VAEs), with decoders better aligned for generation from the prior.
We introduce a novel one-parameter variational objective that lower bounds the data evidence and enables the estimation of approximate fractional posteriors. We extend this framework to hierarchical construction and Bayes posteriors, offering a versatile tool for probabilistic modelling. We demonstrate two cases where gradients can be obtained analytically and a simulation study on mixture models showing that our fractional posteriors can be used to achieve better calibration compared to posteriors from the conventional variational bound. When applied to variational autoencoders (VAEs), our approach attains higher evidence bounds and enables learning of high-performing approximate Bayes posteriors jointly with fractional posteriors. We show that VAEs trained with fractional posteriors produce decoders that are better aligned for generation from the prior.