ePrints

Informing deep neural networks by multiscale principles of neuromodulatory systems

Lookup NU author(s): Dr Srikanth RamaswamyORCiD

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.

Abstract

© 2022 Elsevier Ltd. Our brains have evolved the ability to configure and adapt their processing states to match the unique challenges of acting and learning in diverse environments and behavioral contexts. In biological nervous systems, such state specification and adaptation arise in part from neuromodulators, including acetylcholine, noradrenaline, serotonin, and dopamine, whose diffuse release fine-tunes neuronal and synaptic dynamics and plasticity to complement the behavioral context in real-time. Despite the demonstrated effectiveness of deep neural networks for specific tasks, they remain relatively inflexible at generalizing across tasks or adapting to ever-changing behavioral demands. In this article, we provide an overview of neuromodulatory systems and their relationship to emerging pertinent principles in deep neural networks. We further outline opportunities for the integration of neuromodulatory principles into deep neural networks, towards endowing artificial intelligence with a key ingredient underlying the flexibility and learning capability of biological systems.

Publication metadata

Author(s): Mei J, Muller E, Ramaswamy S

Publication type: Review

Publication status: Published

Journal: Trends in Neurosciences

Year: 2022

Volume: 45

Issue: 3

Pages: 237-250

Print publication date: 01/03/2022

Online publication date: 21/01/2022

Acceptance date: 02/04/2018

ISSN (print): 0166-2236

ISSN (electronic): 1878-108X

Publisher: Elsevier Ltd

URL: https://doi.org/10.1016/j.tins.2021.12.008

DOI: 10.1016/j.tins.2021.12.008

PubMed id: 35074219

Share