Cortical microcircuits as gated-recurrent neural networks

TL;DR

This paper introduces a biologically inspired recurrent neural network model called subLSTM, which mimics cortical microcircuits and performs comparably to traditional LSTMs on sequence tasks, bridging neuroscience and machine learning.

Contribution

The paper proposes a new gated-recurrent neural network model, subLSTM, that maps onto cortical microcircuits, offering a biologically plausible alternative to LSTMs with similar performance.

Findings

subLSTM achieves comparable performance to LSTM in sequence tasks

subLSTM provides a biologically plausible model of cortical microcircuits

the model offers insights into cortical computation and recurrent network design

Abstract

Cortical circuits exhibit intricate recurrent architectures that are remarkably similar across different brain areas. Such stereotyped structure suggests the existence of common computational principles. However, such principles have remained largely elusive. Inspired by gated-memory networks, namely long short-term memory networks (LSTMs), we introduce a recurrent neural network in which information is gated through inhibitory cells that are subtractive (subLSTM). We propose a natural mapping of subLSTMs onto known canonical excitatory-inhibitory cortical microcircuits. Our empirical evaluation across sequential image classification and language modelling tasks shows that subLSTM units can achieve similar performance to LSTM units. These results suggest that cortical circuits can be optimised to solve complex contextual problems and proposes a novel view on their computational function. Overall our work provides a step towards unifying recurrent networks as used in machine learning with their biological counterparts.