Effectiveness of the Recent Advances in Capsule Networks

TL;DR

This paper reviews recent advances in capsule networks, highlighting architectural and routing improvements, analyzing the impact of squash functions, and discussing future research opportunities in the field.

Contribution

It provides a comprehensive overview of recent capsule network developments and offers new insights into the role of squash functions in their performance.

Findings

Focus on routing mechanisms and architecture modifications in recent literature.

Finesse components like squash functions are under-studied.

Insights into how squash functions affect capsule network performance.

Abstract

Convolutional neural networks (CNNs) have revolutionized the field of deep neural networks. However, recent research has shown that CNNs fail to generalize under various conditions and hence the idea of capsules was introduced in 2011, though the real surge of research started from 2017. In this paper, we present an overview of the recent advances in capsule architecture and routing mechanisms. In addition, we find that the relative focus in recent literature is on modifying routing procedure or architecture as a whole but the study of other finer components, specifically, squash function is wanting. Thus, we also present some new insights regarding the effect of squash functions in performance of the capsule networks. Finally, we conclude by discussing and proposing possible opportunities in the field of capsule networks.