# Modularity as a Means for Complexity Management in Neural Networks Learning

arXiv: https://arxiv.org/abs/1902.09240 · HTML: https://mlsanity.com/paper/1902.09240/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Feb 2019.

## Built on by (most cited first, 1 of 1)

- [Towards Understanding the Link Between Modularity and Performance in Neural Networks for Reinforcement Learning](https://mlsanity.com/paper/2205.06451/) (2022)

## Abstract (arXiv)

Training a Neural Network (NN) with lots of parameters or intricate architectures creates undesired phenomena that complicate the optimization process. To address this issue we propose a first modular approach to NN design, wherein the NN is decomposed into a control module and several functional modules, implementing primitive operations. We illustrate the modular concept by comparing performances between a monolithic and a modular NN on a list sorting problem and show the benefits in terms of training speed, training stability and maintainability. We also discuss some questions that arise in modular NNs.