An Intrinsic Value System for Developing Multiple Invariant Representations with Incremental Slowness Learning

Luciw, Matthew and Kompella, Varun and Kazerounian, Sohrob and Schmidhuber, Juergen (2013) An Intrinsic Value System for Developing Multiple Invariant Representations with Incremental Slowness Learning. Frontiers in Neurorobotics, 7 (9). ISSN 1662-5218

Text fin13-luciwkomp.pdf Download (5MB)

Abstract

Curiosity Driven Modular Incremental Slow Feature Analysis (CD-MISFA; [21]) is a recently introduced model of intrinsically-motivated invariance learning. Artificial curiosity enables the orderly formation of multiple stable sensory representations to simplify the agent’s complex sensory input. We discuss computational properties of the CD-MISFA model itself as well as neurophysiological analogs fulfilling similar functional roles. CD-MISFA combines 1. unsupervised representation learning through the slowness principle, 2. generation of an intrinsic reward signal through learning progress of the developing features, and 3. balancing of exploration and exploitation to maximize learning progress and quickly learn multiple feature sets for perceptual simplification. Experimental results on synthetic observations and on the iCub robot show that the intrinsic value system is essential for representation learning. Representations are typically explored and learned in order from least to most costly, as predicted by the theory of curiosity.

Item Type:	Scientific journal article, Newspaper article or Magazine article
Subjects:	Computer sciences > Artificial intelligence > Artificial intelligence not elsewhere classified
Depositing User:	Matthew Luciw
Date Deposited:	13 Mar 2014 13:52
Last Modified:	23 May 2016 08:53
URI:	http://repository.supsi.ch/id/eprint/3698

Actions (login required)

View Item