Development of proactive and reactive behavior via meta-learning of prediction error variance

Shingo Murata, Jun Namikawa, Hiroaki Arie, Jun Tani, Shigeki Sugano

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper investigates a possible neurodynamic mechanism that enables autonomous switching between two basic behavioral modes, namely a "proactive mode" and a "reactive mode." In the proactive mode, actions are generated as intended, whereas in the reactive mode actions are generated in response to the sensory state.We conducted neurorobotics experiments to investigate how these two modes can develop and how a robot can learn to switch autonomously between the two modes as necessary by utilizing our recently developed dynamic neural network model. Tasks designed for the robot included switching between proactive imitation of other's predictable movements using acquired memories and reactive following of other's unpredictable movements through iterative learning of alternating predictable and unpredictable movement patterns. The experimental results revealed that this "meta-learning" capability can lead to self-organization of adequate contextual dynamical structures that can perform autonomous switching between the different behavioral modes.

Original languageEnglish
Title of host publicationNeural Information Processing - 20th International Conference, ICONIP 2013, Proceedings
Pages537-544
Number of pages8
EditionPART 1
DOIs
Publication statusPublished - 2013 Dec 1
Event20th International Conference on Neural Information Processing, ICONIP 2013 - Daegu, Korea, Republic of
Duration: 2013 Nov 32013 Nov 7

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume8226 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other20th International Conference on Neural Information Processing, ICONIP 2013
CountryKorea, Republic of
CityDaegu
Period13/11/313/11/7

    Fingerprint

Keywords

  • Humanoid robot
  • Neurorobotics
  • Recurrent neural network

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Murata, S., Namikawa, J., Arie, H., Tani, J., & Sugano, S. (2013). Development of proactive and reactive behavior via meta-learning of prediction error variance. In Neural Information Processing - 20th International Conference, ICONIP 2013, Proceedings (PART 1 ed., pp. 537-544). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8226 LNCS, No. PART 1). https://doi.org/10.1007/978-3-642-42054-2_67