Attentive Group Equivariant Convolutional Networks

TL;DR

This paper introduces attentive group equivariant convolutions that incorporate attention mechanisms into group convolutions, enhancing the learning of symmetry relationships and improving performance on image benchmarks.

Contribution

It generalizes group convolutions by integrating attention, allowing the network to focus on meaningful symmetry combinations and providing interpretability through attention map visualizations.

Findings

Outperforms traditional group convolutional networks on benchmark datasets

Provides interpretability via visualization of equivariant attention maps

Unifies prior visual attention methods as special cases

Abstract

Although group convolutional networks are able to learn powerful representations based on symmetry patterns, they lack explicit means to learn meaningful relationships among them (e.g., relative positions and poses). In this paper, we present attentive group equivariant convolutions, a generalization of the group convolution, in which attention is applied during the course of convolution to accentuate meaningful symmetry combinations and suppress non-plausible, misleading ones. We indicate that prior work on visual attention can be described as special cases of our proposed framework and show empirically that our attentive group equivariant convolutional networks consistently outperform conventional group convolutional networks on benchmark image datasets. Simultaneously, we provide interpretability to the learned concepts through the visualization of equivariant attention maps.