Evenly Cascaded Convolutional Networks

TL;DR

This paper introduces Evenly Cascaded Convolutional Networks (ECN), a novel neural network architecture inspired by wavelet analysis that features evenly structured, interpretable feature maps and achieves state-of-the-art results on small networks.

Contribution

ECN's evenly structured design simplifies training, produces interpretable features, and achieves new state-of-the-art results for small neural networks without additional pruning.

Findings

ECN achieves 95.24% accuracy on CIFAR-10 with under 500k parameters.

ECN outperforms existing small networks on CIFAR-100.

A 3 million parameter ECN is competitive with state-of-the-art models.

Abstract

We introduce Evenly Cascaded convolutional Network (ECN), a neural network taking inspiration from the cascade algorithm of wavelet analysis. ECN employs two feature streams - a low-level and high-level steam. At each layer these streams interact, such that low-level features are modulated using advanced perspectives from the high-level stream. ECN is evenly structured through resizing feature map dimensions by a consistent ratio, which removes the burden of ad-hoc specification of feature map dimensions. ECN produces easily interpretable features maps, a result whose intuition can be understood in the context of scale-space theory. We demonstrate that ECN's design facilitates the training process through providing easily trainable shortcuts. We report new state-of-the-art results for small networks, without the need for additional treatment such as pruning or compression - a consequence of ECN's simple structure and direct training. A 6-layered ECN design with under 500k parameters achieves 95.24% and 78.99% accuracy on CIFAR-10 and CIFAR-100 datasets, respectively, outperforming the current state-of-the-art on small parameter networks, and a 3 million parameter ECN produces results competitive to the state-of-the-art.