Comparison of Methods Generalizing Max- and Average-Pooling

TL;DR

This paper compares various generalized pooling methods in CNNs, introduces a smooth maximum approximation, and finds that these advanced methods do not outperform standard max- or average-pooling in image classification.

Contribution

It provides a systematic comparison of generalized pooling methods and introduces a new smooth maximum pooling approach.

Findings

No significant performance improvement over standard pooling methods

Proposed smooth maximum pooling offers a theoretical alternative

Empirical evaluation on large-scale image dataset

Abstract

Max- and average-pooling are the most popular pooling methods for downsampling in convolutional neural networks. In this paper, we compare different pooling methods that generalize both max- and average-pooling. Furthermore, we propose another method based on a smooth approximation of the maximum function and put it into context with related methods. For the comparison, we use a VGG16 image classification network and train it on a large dataset of natural high-resolution images (Google Open Images v5). The results show that none of the more sophisticated methods perform significantly better in this classification task than standard max- or average-pooling.