Cargando…
Introducing neuromodulation in deep neural networks to learn adaptive behaviours
Animals excel at adapting their intentions, attention, and actions to the environment, making them remarkably efficient at interacting with a rich, unpredictable and ever-changing external world, a property that intelligent machines currently lack. Such an adaptation property relies heavily on cellu...
Autores principales: | , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Public Library of Science
2020
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6984695/ https://www.ncbi.nlm.nih.gov/pubmed/31986189 http://dx.doi.org/10.1371/journal.pone.0227922 |
_version_ | 1783491677141336064 |
---|---|
author | Vecoven, Nicolas Ernst, Damien Wehenkel, Antoine Drion, Guillaume |
author_facet | Vecoven, Nicolas Ernst, Damien Wehenkel, Antoine Drion, Guillaume |
author_sort | Vecoven, Nicolas |
collection | PubMed |
description | Animals excel at adapting their intentions, attention, and actions to the environment, making them remarkably efficient at interacting with a rich, unpredictable and ever-changing external world, a property that intelligent machines currently lack. Such an adaptation property relies heavily on cellular neuromodulation, the biological mechanism that dynamically controls intrinsic properties of neurons and their response to external stimuli in a context-dependent manner. In this paper, we take inspiration from cellular neuromodulation to construct a new deep neural network architecture that is specifically designed to learn adaptive behaviours. The network adaptation capabilities are tested on navigation benchmarks in a meta-reinforcement learning context and compared with state-of-the-art approaches. Results show that neuromodulation is capable of adapting an agent to different tasks and that neuromodulation-based approaches provide a promising way of improving adaptation of artificial systems. |
format | Online Article Text |
id | pubmed-6984695 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2020 |
publisher | Public Library of Science |
record_format | MEDLINE/PubMed |
spelling | pubmed-69846952020-02-07 Introducing neuromodulation in deep neural networks to learn adaptive behaviours Vecoven, Nicolas Ernst, Damien Wehenkel, Antoine Drion, Guillaume PLoS One Research Article Animals excel at adapting their intentions, attention, and actions to the environment, making them remarkably efficient at interacting with a rich, unpredictable and ever-changing external world, a property that intelligent machines currently lack. Such an adaptation property relies heavily on cellular neuromodulation, the biological mechanism that dynamically controls intrinsic properties of neurons and their response to external stimuli in a context-dependent manner. In this paper, we take inspiration from cellular neuromodulation to construct a new deep neural network architecture that is specifically designed to learn adaptive behaviours. The network adaptation capabilities are tested on navigation benchmarks in a meta-reinforcement learning context and compared with state-of-the-art approaches. Results show that neuromodulation is capable of adapting an agent to different tasks and that neuromodulation-based approaches provide a promising way of improving adaptation of artificial systems. Public Library of Science 2020-01-27 /pmc/articles/PMC6984695/ /pubmed/31986189 http://dx.doi.org/10.1371/journal.pone.0227922 Text en © 2020 Vecoven et al http://creativecommons.org/licenses/by/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. |
spellingShingle | Research Article Vecoven, Nicolas Ernst, Damien Wehenkel, Antoine Drion, Guillaume Introducing neuromodulation in deep neural networks to learn adaptive behaviours |
title | Introducing neuromodulation in deep neural networks to learn adaptive behaviours |
title_full | Introducing neuromodulation in deep neural networks to learn adaptive behaviours |
title_fullStr | Introducing neuromodulation in deep neural networks to learn adaptive behaviours |
title_full_unstemmed | Introducing neuromodulation in deep neural networks to learn adaptive behaviours |
title_short | Introducing neuromodulation in deep neural networks to learn adaptive behaviours |
title_sort | introducing neuromodulation in deep neural networks to learn adaptive behaviours |
topic | Research Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6984695/ https://www.ncbi.nlm.nih.gov/pubmed/31986189 http://dx.doi.org/10.1371/journal.pone.0227922 |
work_keys_str_mv | AT vecovennicolas introducingneuromodulationindeepneuralnetworkstolearnadaptivebehaviours AT ernstdamien introducingneuromodulationindeepneuralnetworkstolearnadaptivebehaviours AT wehenkelantoine introducingneuromodulationindeepneuralnetworkstolearnadaptivebehaviours AT drionguillaume introducingneuromodulationindeepneuralnetworkstolearnadaptivebehaviours |