# The Multi-Lane Capsule Network (MLCN)

## TL;DR

The paper presents Multi-Lane Capsule Networks (MLCN), a resource-efficient parallel architecture that maintains high accuracy and speeds up training and inference compared to traditional Capsule Networks.

## Contribution

MLCN introduces a parallel lane organization for Capsule Networks, reducing parameters and computational cost while improving speed and maintaining accuracy.

## Key findings

- Achieves similar accuracy to CapsNet with fewer parameters.
- Outperforms original CapsNet with a novel lane configuration.
- More than twice as fast in training and inference.

## Abstract

We introduce Multi-Lane Capsule Networks (MLCN), which are a separable and resource efficient organization of Capsule Networks (CapsNet) that allows parallel processing, while achieving high accuracy at reduced cost. A MLCN is composed of a number of (distinct) parallel lanes, each contributing to a dimension of the result, trained using the routing-by-agreement organization of CapsNet. Our results indicate similar accuracy with a much reduced cost in number of parameters for the Fashion-MNIST and Cifar10 datsets. They also indicate that the MLCN outperforms the original CapsNet when using a proposed novel configuration for the lanes. MLCN also has faster training and inference times, being more than two-fold faster than the original CapsNet in the same accelerator.