Convolutional Conditional Neural Processes

Jonathan Gordon, Wessel P. Bruinsma, Andrew Y. K. Foong, James Requeima, Yann Dubois, Richard E. Turner

Keywords: equivariance, generalization, inductive bias

Tuesday: Feature Discovery for Structured Data

Abstract: We introduce the Convolutional Conditional Neural Process (ConvCNP), a new member of the Neural Process family that models translation equivariance in the data. Translation equivariance is an important inductive bias for many learning problems including time series modelling, spatial data, and images. The model embeds data sets into an infinite-dimensional function space, as opposed to finite-dimensional vector spaces. To formalize this notion, we extend the theory of neural representations of sets to include functional representations, and demonstrate that any translation-equivariant embedding can be represented using a convolutional deep-set. We evaluate ConvCNPs in several settings, demonstrating that they achieve state-of-the-art performance compared to existing NPs. We demonstrate that building in translation equivariance enables zero-shot generalization to challenging, out-of-domain tasks.

Similar Papers

Scale-Equivariant Steerable Networks
Ivan Sosnovik, Michał Szmaja, Arnold Smeulders,
Building Deep Equivariant Capsule Networks
Sai Raam Venkataraman, S. Balasubramanian, R. Raghunatha Sarma,