Interactional Motivation in Artificial Systems: Between Extrinsic and Intrinsic Motivation

Olivier Georgeon 1 James Marshall 2 Simon Gay 1
1 SILEX - Supporting Interaction and Learning by Experience
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
2 Sarah Lawrence College
Abstract : This paper introduces Interactional Motivation (IM) as a way to implement self-motivation in artificial systems. An interactionally motivated agent selects behaviors for the sake of enacting the behavior itself rather than for the value of the behavior’s outcome. IM contrasts with extrinsic motivation by the fact that it defines the agent’s motivation independently from the environment’s state. Because IM does not refer to the environment’s states, we argue that IM is a form of self-motivation on the same level as intrinsic motivation. IM, however, differs from intrinsic motivation by the fact that IM allows specifying the agent’s inborn value system explicitly. This paper introduces a formal definition of the IM paradigm and compares it to the reinforcement-learning paradigm as traditionally implemented in Partially Observable Markov Decision Processes (POMDPs).
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [10 references]  Display  Hide  Download
https://hal.archives-ouvertes.fr/hal-01353128
Contributor : Équipe Gestionnaire Des Publications Si Liris <>
Submitted on : Thursday, October 13, 2016 - 4:28:55 PM
Last modification on : Tuesday, April 9, 2019 - 3:36:04 PM
Document(s) archivé(s) le : Saturday, February 4, 2017 - 10:02:01 PM

File

PID2566165.pdf
Files produced by the author(s)

Identifiers

Collections

EC-LYON | LIRIS | INSA-LYON | UNIV-LYON1 | INSA-GROUPE

Citation

Olivier Georgeon, James Marshall, Simon Gay. Interactional Motivation in Artificial Systems: Between Extrinsic and Intrinsic Motivation. International Conference on Development and Learning (ICDL-Epirob), Nov 2012, San Diego, CA, United States. pp.1-2, ⟨10.1109/DevLrn.2012.6400833⟩. ⟨hal-01353128⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

224

Files downloads

272

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD