Perforated Backpropagation: A Neuroscience Inspired Extension to Artificial Neural Networks

TL;DR

This paper introduces Perforated Backpropagation, a biologically inspired modification to neural network training that enhances performance and enables model compression by incorporating dendrite-like nodes trained to reduce error.

Contribution

It proposes a novel training method inspired by biological dendrites, adding and training Dendrite Nodes to improve accuracy and compression in deep neural networks.

Findings

Improved accuracy across multiple domains.

Achieved significant model compression without accuracy loss.

Enhanced training efficiency with biologically inspired architecture.

Abstract

The neurons of artificial neural networks were originally invented when much less was known about biological neurons than is known today. Our work explores a modification to the core neuron unit to make it more parallel to a biological neuron. The modification is made with the knowledge that biological dendrites are not simply passive activation funnels, but also compute complex non-linear functions as they transmit activation to the cell body. The paper explores a novel system of "Perforated" backpropagation empowering the artificial neurons of deep neural networks to achieve better performance coding for the same features they coded for in the original architecture. After an initial network training phase, additional "Dendrite Nodes" are added to the network and separately trained with a different objective: to correlate their output with the remaining error of the original neurons. The trained Dendrite Nodes are then frozen, and the original neurons are further trained, now taking into account the additional error signals provided by the Dendrite Nodes. The cycle of training the original neurons and then adding and training Dendrite Nodes can be repeated several times until satisfactory performance is achieved. Our algorithm was successfully added to modern state-of-the-art PyTorch networks across multiple domains, improving upon original accuracies and allowing for significant model compression without a loss in accuracy.