Sorted Pooling in Convolutional Networks for One-shot Learning

TL;DR

This paper introduces sorted pooling methods for convolutional networks that enhance generalization and accuracy, especially in data-scarce one-shot learning scenarios, by selecting specific response responses within pooling regions.

Contribution

It proposes generalized pooling operations, including the $k$th maximum and sorted pooling, to improve network performance in limited data settings.

Findings

Improved accuracy in one-shot learning tasks.

Reduced training time and error rates.

Enhanced generalization capabilities.

Abstract

We present generalized versions of the commonly used maximum pooling operation: $k$th maximum and sorted pooling operations which selects the $k$th largest response in each pooling region, selecting locally consistent features of the input images. This method is able to increase the generalization power of a network and can be used to decrease training time and error rate of networks and it can significantly improve accuracy in case of training scenarios where the amount of available data is limited, like one-shot learning scenarios