Curiosity: Emergent Behavior Through Interacting Multi-Level Predictions
Document Type
Poster Session
Publication Date
5-7-2017
Published In
Designing For Curiosity: An Interdisciplinary Workshop
Abstract
Over the past 15 years our research group has been exploring models of developmental robotics and curiosity. Our research is based on the premise that intelligent behavior arises through emergent interactions between opposing forces in an open-ended, task-independent environment. In an initial experiment we constructed a recurrent neural network model where self-motivation was "an emergent property generated by the competing pressures that arise in attempting to balance predictability and novelty". The system first focused on its error, then learned to successfully predict its error, and finally became habituated to what caused the error. This process of focusing, learning, and habituating can be seen as a rudimentary type of curiosity.
Conference
Designing For Curiosity: An Interdisciplinary Workshop
Conference Dates
May 7, 2017
Conference Location
Denver, CO
Recommended Citation
Lisa A. Meeden, D. Blank, and J. Marshall.
(2017).
"Curiosity: Emergent Behavior Through Interacting Multi-Level Predictions".
Designing For Curiosity: An Interdisciplinary Workshop.
https://works.swarthmore.edu/fac-comp-sci/85