Implicit Equivariance in Convolutional Networks

Naman Khetan, Tushar Arora, Suraj Sharan, Samee Ur Rehman, Deepak Gupta

Abstract: Convolutional Neural Networks (CNN) are inherently equivariant under translations, however, they do not have an equivalent embedded mechanism to handle other transformations such as rotations and change in scale. Several approaches have been proposed that make CNNs equivariant under other transformation groups by design. Among these, steerable CNNs have been especially effective. However, these approaches require redesigning standard networks with filters mapped from combinations of predefined basis involving complex analytical functions. We experimentally demonstrate that these restrictions in the choice of basis can lead to model weights that are sub-optimal for the primary deep learning task (e.g. classification). Moreover, such hard-baked explicit formulations make it difficult to design composite networks comprising heterogeneous feature groups. To circumvent such issues, we propose Implicitly Equivariant Networks (IEN) which induce equivariance in the different layers of a standard CNN model by optimizing a multi-objective loss function that combines the primary loss with an equivariance loss term. Through experiments with VGG and ResNet models on Rot-MNIST , Rot-TinyImageNet, Scale-MNIST and STL-10 datasets, we show that IEN, even with its simple formulation, performs better than steerable networks. Also, IEN facilitates construction of heterogeneous filter groups allowing to reduce channels in CNNs by factors of over 30% while maintaining performance at par with baselines. The efficacy of IEN is further validated on the hard problem of visual object tracking. We show that IEN outperforms state-of-the-art rotation equivariant tracking method while providing faster inference speed.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2111.14157/code)