Quantifying Emergent Behavior of Autonomous Robots

TL;DR

This paper introduces a novel approach to quantify the complexity of autonomous robot behaviors using a new decomposition of excess entropy, accounting for resolution dependence, and demonstrates how behavior complexity evolves with learning.

Contribution

It proposes a new decomposition of excess entropy into resolution-dependent and independent parts and introduces a correlation integral-based estimation method for continuous systems.

Findings

Behavior complexity increases with learning duration.

The proposed method effectively quantifies nonrandom structure in robot behaviors.

Resolution dependence of complexity measures is addressed and analyzed.

Abstract

Quantifying behaviors of robots which were generated autonomously from task-independent objective functions is an important prerequisite for objective comparisons of algorithms and movements of animals. The temporal sequence of such a behavior can be considered as a time series and hence complexity measures developed for time series are natural candidates for its quantification. The predictive information and the excess entropy are such complexity measures. They measure the amount of information the past contains about the future and thus quantify the nonrandom structure in the temporal sequence. However, when using these measures for systems with continuous states one has to deal with the fact that their values will depend on the resolution with which the systems states are observed. For deterministic systems both measures will diverge with increasing resolution. We therefore propose a new decomposition of the excess entropy in resolution dependent and resolution independent parts and discuss how they depend on the dimensionality of the dynamics, correlations and the noise level. For the practical estimation we propose to use estimates based on the correlation integral instead of the direct estimation of the mutual information using the algorithm by Kraskov et al. (2004) which is based on next neighbor statistics because the latter allows less control of the scale dependencies. Using our algorithm we are able to show how autonomous learning generates behavior of increasing complexity with increasing learning duration.