Deeply Shared Filter Bases for Parameter-Efficient Convolutional Neural Networks

TL;DR

This paper introduces a recursive convolutional block with a learnable filter basis and orthogonality regularization, enabling parameter sharing in CNNs without performance loss and improving training stability and representation capacity.

Contribution

It proposes a novel recursive convolution block design with a learnable filter basis and orthogonality regularization to enhance parameter sharing in CNNs.

Findings

Outperforms overparameterized networks in image classification and object detection.

Orthogonality regularization improves gradient flow during training.

Parameter sharing does not degrade performance, thanks to the proposed design.

Abstract

Modern convolutional neural networks (CNNs) have massive identical convolution blocks, and, hence, recursive sharing of parameters across these blocks has been proposed to reduce the amount of parameters. However, naive sharing of parameters poses many challenges such as limited representational power and the vanishing/exploding gradients problem of recursively shared parameters. In this paper, we present a recursive convolution block design and training method, in which a recursively shareable part, or a filter basis, is separated and learned while effectively avoiding the vanishing/exploding gradients problem during training. We show that the unwieldy vanishing/exploding gradients problem can be controlled by enforcing the elements of the filter basis orthonormal, and empirically demonstrate that the proposed orthogonality regularization improves the flow of gradients during training. Experimental results on image classification and object detection show that our approach, unlike previous parameter-sharing approaches, does not trade performance to save parameters and consistently outperforms overparameterized counterpart networks. This superior performance demonstrates that the proposed recursive convolution block design and the orthogonality regularization not only prevent performance degradation, but also consistently improve the representation capability while a significant amount of parameters are recursively shared.