Bootstrapping autonomous skill learning in the MDB cognitive architecture

  1. Alejandro Romero 1
  2. Francisco Bellas 1
  3. Jose A. Becerra 1
  4. Richard J. Duro 1
  1. 1 Universidade da Coruña. A Coruña, Spain
Book:
Understanding the Brain Function and Emotions: 8th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2019 Almería, Spain, June 3–7, 2019 Proceedings, Part I
  1. José Manuel Ferrández Vicente (dir. congr.)
  2. José Ramón Álvarez-Sánchez (dir. congr.)
  3. Félix de la Paz López (dir. congr.)
  4. Javier Toledo Moreo (dir. congr.)
  5. Hojjat Adeli (dir. congr.)

Publisher: Springer Suiza

ISBN: 978-3-030-19591-5

Year of publication: 2019

Pages: 120-129

Type: Book chapter

DIALNET GOOGLE SCHOLAR

Abstract

This paper is concerned with motivation in autonomousrobots. In particular we focus on the basic structure that is necessary for bootstrapping the initial stages of multiple skill learning within the motivational engine of the MDB cognitive architecture. To this end, taking inspiration from a series of computational models of the use of motivations in infants, we propose an approach that leverages two types of cognitive motivations: exploratory and proficiency based. The latter modulated by the concept of interestingness. We postulate that these make up the minimum set of motivational components required to initiate the unrewarded learning of a skill toolbox that may later be used in order to achieve operational goals.The approach is illustrated through an experiment with a real robot that is learning skills in a real environment.

Data source: Dialnet