CrescendoNet: A Simple Deep Convolutional Neural Network with Ensemble Behavior

TL;DR

CrescendoNet is a deep convolutional neural network that achieves high performance without residual connections by stacking simple blocks with independent paths, exhibiting ensemble-like behavior and enabling efficient path-wise training.

Contribution

This paper introduces CrescendoNet, a novel deep CNN architecture with linear growth in parameters and layers, outperforming many existing models without residual connections.

Findings

CrescendoNet with 15 layers outperforms similar non-residual networks on CIFAR and SVHN.

It matches DenseNet-BC performance with fewer layers and parameters.

The network's ensemble behavior contributes to its high accuracy.

Abstract

We introduce a new deep convolutional neural network, CrescendoNet, by stacking simple building blocks without residual connections. Each Crescendo block contains independent convolution paths with increased depths. The numbers of convolution layers and parameters are only increased linearly in Crescendo blocks. In experiments, CrescendoNet with only 15 layers outperforms almost all networks without residual connections on benchmark datasets, CIFAR10, CIFAR100, and SVHN. Given sufficient amount of data as in SVHN dataset, CrescendoNet with 15 layers and 4.1M parameters can match the performance of DenseNet-BC with 250 layers and 15.3M parameters. CrescendoNet provides a new way to construct high performance deep convolutional neural networks without residual connections. Moreover, through investigating the behavior and performance of subnetworks in CrescendoNet, we note that the high performance of CrescendoNet may come from its implicit ensemble behavior, which differs from the FractalNet that is also a deep convolutional neural network without residual connections. Furthermore, the independence between paths in CrescendoNet allows us to introduce a new path-wise training procedure, which can reduce the memory needed for training.