Neural Networks are Surprisingly Modular

by   Daniel Filan, et al.

The learned weights of a neural network are often considered devoid of scrutable internal structure. In order to attempt to discern structure in these weights, we introduce a measurable notion of modularity for multi-layer perceptrons (MLPs), and investigate the modular structure of MLPs trained on datasets of small images. Our notion of modularity comes from the graph clustering literature: a "module" is a set of neurons with strong internal connectivity but weak external connectivity. We find that MLPs that undergo training and weight pruning are often significantly more modular than random networks with the same distribution of weights. Interestingly, they are much more modular when trained with dropout. Further analysis shows that this modularity seems to arise mostly for networks trained on learnable datasets. We also present exploratory analyses of the importance of different modules for performance and how modules depend on each other. Understanding the modular structure of neural networks, when such structure exists, will hopefully render their inner workings more interpretable to engineers.


Clusterability in Neural Networks

The learned weights of a neural network have often been considered devoi...

Seeing is Believing: Brain-Inspired Modular Training for Mechanistic Interpretability

We introduce Brain-Inspired Modular Training (BIMT), a method for making...

Independent Modular Networks

Monolithic neural networks that make use of a single set of weights to l...

Modular WSS-based OXCs for Large-Scale Optical Networks

The explosive growth of broadband applications calls for large-scale opt...

Modular Grammatical Evolution for the Generation of Artificial Neural Networks

This paper presents a novel method, called Modular Grammatical Evolution...

Modular meta-learning

Many prediction problems, such as those that arise in the context of rob...

Detecting Modularity in Deep Neural Networks

A neural network is modular to the extent that parts of its computationa...

Code Repositories

Please sign up or login with your details

Forgot password? Click here to reset