Timezone: »

 
Poster
Group Equivariant Subsampling
Jin Xu · Hyunjik Kim · Thomas Rainforth · Yee Teh

Thu Dec 09 08:30 AM -- 10:00 AM (PST) @ Virtual
in Poster Session 6 »

Subsampling is used in convolutional neural networks (CNNs) in the form of pooling or strided convolutions, to reduce the spatial dimensions of feature maps and to allow the receptive fields to grow exponentially with depth. However, it is known that such subsampling operations are not translation equivariant, unlike convolutions that are translation equivariant. Here, we first introduce translation equivariant subsampling/upsampling layers that can be used to construct exact translation equivariant CNNs. We then generalise these layers beyond translations to general groups, thus proposing group equivariant subsampling/upsampling. We use these layers to construct group equivariant autoencoders (GAEs) that allow us to learn low-dimensional equivariant representations. We empirically verify on images that the representations are indeed equivariant to input translations and rotations, and thus generalise well to unseen positions and orientations. We further use GAEs in models that learn object-centric representations on multi-object datasets, and show improved data efficiency and decomposition compared to non-equivariant baselines.

Keywords: Deep Learning · Representation Learning

Author Information

Jin Xu (University of Oxford)
Hyunjik Kim (DeepMind)
Thomas Rainforth (University of Oxford)
Yee Teh (DeepMind)

More from the Same Authors