Document Detail


Modelling brain emergent behaviours through coevolution of neural agents.
MedLine Citation:
PMID:  15990275     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
Recently, many research efforts focus on modelling partial brain areas with the long-term goal to support cognitive abilities of artificial organisms. Existing models usually suffer from heterogeneity, which constitutes their integration very difficult. The present work introduces a computational framework to address brain modelling tasks, emphasizing on the integrative performance of substructures. Moreover, implemented models are embedded in a robotic platform to support its behavioural capabilities. We follow an agent-based approach in the design of substructures to support the autonomy of partial brain structures. Agents are formulated to allow the emergence of a desired behaviour after a certain amount of interaction with the environment. An appropriate collaborative coevolutionary algorithm, able to emphasize both the speciality of brain areas and their cooperative performance, is employed to support design specification of agent structures. The effectiveness of the proposed approach is illustrated through the implementation of computational models for motor cortex and hippocampus, which are successfully tested on a simulated mobile robot.
Authors:
Michail Maniadakis; Panos Trahanias
Related Documents :
18058035 - Should comprehensive diagnosis include idiographic understanding?
20015855 - The association between choice stepping reaction time and falls in older adults--a path...
11162295 - Effects of social-psychological factors on creative performance: the role of informatio...
12766435 - Response to shaffer (1996): the case for a 'complex systems' conceptualisation of addic...
7729385 - Estimation of variation among individuals of biological half-time of cadmium calculated...
24324175 - Evidence for soft bounds in ubuntu package sizes and mammalian body masses.
Publication Detail:
Type:  Comparative Study; Journal Article     Date:  2005-06-29
Journal Detail:
Title:  Neural networks : the official journal of the International Neural Network Society     Volume:  19     ISSN:  0893-6080     ISO Abbreviation:  Neural Netw     Publication Date:  2006 Jun 
Date Detail:
Created Date:  2006-07-14     Completed Date:  2006-09-08     Revised Date:  2006-11-15    
Medline Journal Info:
Nlm Unique ID:  8805018     Medline TA:  Neural Netw     Country:  United States    
Other Details:
Languages:  eng     Pagination:  705-20     Citation Subset:  IM    
Affiliation:
Institute of Computer Science, Foundation for Research and Technology-Hellas (FORTH), P.O. Box 1385, Heraklion, 711 10 Crete, Greece. mmaniada@ics.forth.gr
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Animals
Artificial Intelligence
Behavior / physiology*
Brain / cytology*,  physiology
Computer Simulation
Hippocampus / physiology
Humans
Learning / physiology*
Models, Neurological
Motivation
Motor Cortex / physiology
Neural Inhibition / physiology
Neural Networks (Computer)*
Neural Pathways / physiology
Neurons / physiology*

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  A novel information geometric approach to variable selection in MLP networks.
Next Document:  Using ICA for removal of ocular artifacts in EEG recorded from blind subjects.