Learning emotions in virtual environments

Mannes Poel; Hendrikus J.A. op den Akker; Antinus Nijholt; A.J. van Kesteren; A.-J. van Kesteren

Learning emotions in virtual environments

Mannes Poel, Hendrikus J.A. op den Akker, Antinus Nijholt, A.J. van Kesteren, A.-J. van Kesteren

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

73 Downloads (Pure)

Abstract

A modular hybrid neural network architecture, called SHAME, for emotion learning is introduced. The system learns from annotated data how the emotional state is generated and changes due to internal and external stimuli. Part of the modular architecture is domain independent and part must be adapted to the domain under consideration. The generation and learning of emotions is based on the event appraisal model. The architecture is implemented in a prototype consisting of agents trying to survive in a virtual world. An evaluation of this prototype shows that the architecture is capable of generating natural emotions and furthermore that training of the neural network modules in the architecture is computationally feasible. Keywords: hybrid neural systems, emotions, learning, agents.

Original language	Undefined
Title of host publication	Cybernetics and Systems 2002
Editors	R. Trappl
Place of Publication	Vienna
Publisher	Austrian Society for Cybernetic Studies
Pages	751-755
Number of pages	5
ISBN (Print)	3-85206-160-1
Publication status	Published - 2002
Event	16th European Meeting on Cybernetics and Systems Research 2002 - University of Vienna, Vienna, Austria Duration: 2 Apr 2002 → 5 Apr 2002 Conference number: 16

Publication series

Name
Publisher	Austrian Society for Cybernetic Studies

Conference

Conference	16th European Meeting on Cybernetics and Systems Research 2002
Abbreviated title	Cybernetics and Systems
Country/Territory	Austria
City	Vienna
Period	2/04/02 → 5/04/02

Keywords

EWI-6672
METIS-209273
IR-66305
HMI-IA: Intelligent Agents
HMI-CI: Computational Intelligence

Access to Document

Cite this

@inproceedings{321c6cca2276483aa5c7fa3c1371b8cf,

title = "Learning emotions in virtual environments",

abstract = "A modular hybrid neural network architecture, called SHAME, for emotion learning is introduced. The system learns from annotated data how the emotional state is generated and changes due to internal and external stimuli. Part of the modular architecture is domain independent and part must be adapted to the domain under consideration. The generation and learning of emotions is based on the event appraisal model. The architecture is implemented in a prototype consisting of agents trying to survive in a virtual world. An evaluation of this prototype shows that the architecture is capable of generating natural emotions and furthermore that training of the neural network modules in the architecture is computationally feasible. Keywords: hybrid neural systems, emotions, learning, agents.",

keywords = "EWI-6672, METIS-209273, IR-66305, HMI-IA: Intelligent Agents, HMI-CI: Computational Intelligence",

author = "Mannes Poel and {op den Akker}, {Hendrikus J.A.} and Antinus Nijholt and {van Kesteren}, A.J. and {van Kesteren}, A.-J.",

note = "Imported from HMI; 16th European Meeting on Cybernetics and Systems Research 2002 ; Conference date: 02-04-2002 Through 05-04-2002",

year = "2002",

language = "Undefined",

isbn = "3-85206-160-1",

publisher = "Austrian Society for Cybernetic Studies",

pages = "751--755",

editor = "R. Trappl",

booktitle = "Cybernetics and Systems 2002",

}

TY - GEN

T1 - Learning emotions in virtual environments

AU - Poel, Mannes

AU - op den Akker, Hendrikus J.A.

AU - Nijholt, Antinus

AU - van Kesteren, A.J.

AU - van Kesteren, A.-J.

N1 - Conference code: 16

PY - 2002

Y1 - 2002

N2 - A modular hybrid neural network architecture, called SHAME, for emotion learning is introduced. The system learns from annotated data how the emotional state is generated and changes due to internal and external stimuli. Part of the modular architecture is domain independent and part must be adapted to the domain under consideration. The generation and learning of emotions is based on the event appraisal model. The architecture is implemented in a prototype consisting of agents trying to survive in a virtual world. An evaluation of this prototype shows that the architecture is capable of generating natural emotions and furthermore that training of the neural network modules in the architecture is computationally feasible. Keywords: hybrid neural systems, emotions, learning, agents.

AB - A modular hybrid neural network architecture, called SHAME, for emotion learning is introduced. The system learns from annotated data how the emotional state is generated and changes due to internal and external stimuli. Part of the modular architecture is domain independent and part must be adapted to the domain under consideration. The generation and learning of emotions is based on the event appraisal model. The architecture is implemented in a prototype consisting of agents trying to survive in a virtual world. An evaluation of this prototype shows that the architecture is capable of generating natural emotions and furthermore that training of the neural network modules in the architecture is computationally feasible. Keywords: hybrid neural systems, emotions, learning, agents.

KW - EWI-6672

KW - METIS-209273

KW - IR-66305

KW - HMI-IA: Intelligent Agents

KW - HMI-CI: Computational Intelligence

M3 - Conference contribution

SN - 3-85206-160-1

SP - 751

EP - 755

BT - Cybernetics and Systems 2002

A2 - Trappl, R.

PB - Austrian Society for Cybernetic Studies

CY - Vienna

T2 - 16th European Meeting on Cybernetics and Systems Research 2002

Y2 - 2 April 2002 through 5 April 2002

ER -