The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control
The proposed architecture applies the principle of predictive coding and deep learning in a brain-inspired approach to robotic sensorimotor control. It is composed of many layers each of which is a recurrent network. The component networks can be spontaneously active due to the homeokinetic learning...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2020-09-01
|
| Series: | Frontiers in Neurorobotics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fnbot.2020.00062/full |
| _version_ | 1818970941362798592 |
|---|---|
| author | Simón C. Smith Richard Dharmadi Calum Imrie Bailu Si Bailu Si J. Michael Herrmann J. Michael Herrmann |
| author_facet | Simón C. Smith Richard Dharmadi Calum Imrie Bailu Si Bailu Si J. Michael Herrmann J. Michael Herrmann |
| author_sort | Simón C. Smith |
| collection | DOAJ |
| description | The proposed architecture applies the principle of predictive coding and deep learning in a brain-inspired approach to robotic sensorimotor control. It is composed of many layers each of which is a recurrent network. The component networks can be spontaneously active due to the homeokinetic learning rule, a principle that has been studied previously for the purpose of self-organized generation of behavior. We present robotic simulations that illustrate the function of the network and show evidence that deeper networks enable more complex exploratory behavior. |
| first_indexed | 2024-12-20T14:44:29Z |
| format | Article |
| id | doaj.art-5d9066e2dab94b6b9f08ce74eb4d924f |
| institution | Directory Open Access Journal |
| issn | 1662-5218 |
| language | English |
| last_indexed | 2024-12-20T14:44:29Z |
| publishDate | 2020-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurorobotics |
| spelling | doaj.art-5d9066e2dab94b6b9f08ce74eb4d924f2022-12-21T19:37:10ZengFrontiers Media S.A.Frontiers in Neurorobotics1662-52182020-09-011410.3389/fnbot.2020.00062568309The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot ControlSimón C. Smith0Richard Dharmadi1Calum Imrie2Bailu Si3Bailu Si4J. Michael Herrmann5J. Michael Herrmann6Institute of Perception, Action and Behaviour (IPAB), School of Informatics, University of Edinburgh, Edinburgh, United KingdomInstitute of Perception, Action and Behaviour (IPAB), School of Informatics, University of Edinburgh, Edinburgh, United KingdomInstitute of Perception, Action and Behaviour (IPAB), School of Informatics, University of Edinburgh, Edinburgh, United KingdomState Key Laboratory of Robotics, Shenyang Institute of Automation, Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, ChinaSchool of Systems Science, Beijing Normal University, Beijing, ChinaInstitute of Perception, Action and Behaviour (IPAB), School of Informatics, University of Edinburgh, Edinburgh, United KingdomState Key Laboratory of Robotics, Shenyang Institute of Automation, Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, ChinaThe proposed architecture applies the principle of predictive coding and deep learning in a brain-inspired approach to robotic sensorimotor control. It is composed of many layers each of which is a recurrent network. The component networks can be spontaneously active due to the homeokinetic learning rule, a principle that has been studied previously for the purpose of self-organized generation of behavior. We present robotic simulations that illustrate the function of the network and show evidence that deeper networks enable more complex exploratory behavior.https://www.frontiersin.org/article/10.3389/fnbot.2020.00062/fulldeep neural networksautonomous learninghomeokinesisself-organizing controlrobot control |
| spellingShingle | Simón C. Smith Richard Dharmadi Calum Imrie Bailu Si Bailu Si J. Michael Herrmann J. Michael Herrmann The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control Frontiers in Neurorobotics deep neural networks autonomous learning homeokinesis self-organizing control robot control |
| title | The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control |
| title_full | The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control |
| title_fullStr | The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control |
| title_full_unstemmed | The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control |
| title_short | The DIAMOND Model: Deep Recurrent Neural Networks for Self-Organizing Robot Control |
| title_sort | diamond model deep recurrent neural networks for self organizing robot control |
| topic | deep neural networks autonomous learning homeokinesis self-organizing control robot control |
| url | https://www.frontiersin.org/article/10.3389/fnbot.2020.00062/full |
| work_keys_str_mv | AT simoncsmith thediamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT richarddharmadi thediamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT calumimrie thediamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT bailusi thediamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT bailusi thediamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT jmichaelherrmann thediamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT jmichaelherrmann thediamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT simoncsmith diamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT richarddharmadi diamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT calumimrie diamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT bailusi diamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT bailusi diamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT jmichaelherrmann diamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol AT jmichaelherrmann diamondmodeldeeprecurrentneuralnetworksforselforganizingrobotcontrol |