Fingerprinting Agent-Environment Interaction via Information Theory

In this paper, we investigate by means of statistical and information-theoretic measures, to what extent sensory-motor coordinated activity can generate and structure information in the sensory channels of a simulated agent interacting with its surrounding environment. We show how the usage of correlation, entropy, and mutual information can be employed (a) to segment an observed behavior into distinct behavioral states, (b) to quantify (fingerprint) the agent-environment interaction, and (c) to analyze the informational relationship between the different components of the sensory-motor apparatus. We hypothesize that a deeper understanding of the information-theoretic implications of sensory-motor coordination can help us endow our robots with better sensory morphologies, and with better strategies for exploring their surrounding environment.


Published in:
The Eighth International Conference on Intelligent Autonomous Systems (IAS-8), 10, 512-520
Presented at:
The Eighth International Conference on Intelligent Autonomous Systems, Amsterdam, March 10 to 12, 2004
Year:
2004
Keywords:
Laboratories:




 Record created 2006-04-18, last modified 2018-03-17

n/a:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)