Parametric Exponential Linear Unit for Deep Convolutional Neural Networks

TL;DR

This paper introduces a learnable parametric version of the Exponential Linear Unit (ELU) activation function, called PELU, which adapts its shape during training to improve CNN performance across multiple datasets.

Contribution

It proposes a novel parametric ELU that learns the optimal activation shape at each layer, enhancing CNN accuracy with minimal additional parameters.

Findings

PELU outperforms non-parametric ELU on MNIST, CIFAR-10/100, and ImageNet.

Achieves up to 7.28% error reduction on ImageNet with minimal parameter increase.

Networks learn diverse activation behaviors across layers using PELU.

Abstract

Object recognition is an important task for improving the ability of visual systems to perform complex scene understanding. Recently, the Exponential Linear Unit (ELU) has been proposed as a key component for managing bias shift in Convolutional Neural Networks (CNNs), but defines a parameter that must be set by hand. In this paper, we propose learning a parameterization of ELU in order to learn the proper activation shape at each layer in the CNNs. Our results on the MNIST, CIFAR-10/100 and ImageNet datasets using the NiN, Overfeat, All-CNN and ResNet networks indicate that our proposed Parametric ELU (PELU) has better performances than the non-parametric ELU. We have observed as much as a 7.28% relative error improvement on ImageNet with the NiN network, with only 0.0003% parameter increase. Our visual examination of the non-linear behaviors adopted by Vgg using PELU shows that the network took advantage of the added flexibility by learning different activations at different layers.