Timezone: »

 
Learning Generative Models with Invariance to Symmetries
James Allingham · Javier Antorán · Shreyas Padhy · Eric Nalisnick · José Miguel Hernández-Lobato

@
in Symmetry and Geometry in Neural Representations (NeurReps) »
Event URL: https://openreview.net/forum?id=Ff1N3et1IV »

While imbuing a model with invariance to symmetries can improve data efficiency and predictive performance, most methods require specialised architectures and thus prior knowledge of the symmetries. Unfortunately, we don't always know what symmetries are present in the data. Recent work has solved this problem by jointly learning the invariance (or the degree of invariance) with the model from the data alone. But, this work has focused on discriminative models. We describe a method for learning invariant generative models. We demonstrate that our method can learn a generative model of handwritten digits that is invariant to rotation. We hope this line of work will enable more data-efficient deep generative models.

Keywords: Generative Models · Invariance · Symmetry

Author Information

James Allingham (University of Cambridge)
Javier Antorán (University of Cambridge)
Shreyas Padhy (University of Cambridge)
Eric Nalisnick (University of Amsterdam)
José Miguel Hernández-Lobato (University of Cambridge)

More from the Same Authors