Quantifying the Evolutionary Self Structuring of Embodied Cognitive Networks

TL;DR

This paper proposes a theoretical framework for modeling embodied cognitive systems, emphasizing the importance of group structure in motion and stochastic environments, aiming to aid design and understanding of self-organizing agents.

Contribution

It introduces a novel modeling approach that accounts for body motion group structure and stochastic environments, simplifying the design of self-organizing embodied systems.

Findings

Framework emphasizes SE(3) group structure over Euclidean space

Proposes algorithms for informational measures in self-organization

Aims to reduce computational complexity in modeling

Abstract

We outline a possible theoretical framework for the quantitative modeling of networked embodied cognitive systems. We notice that: 1) information self structuring through sensory-motor coordination does not deterministically occur in Rn vector space, a generic multivariable space, but in SE(3), the group structure of the possible motions of a body in space; 2) it happens in a stochastic open ended environment. These observations may simplify, at the price of a certain abstraction, the modeling and the design of self organization processes based on the maximization of some informational measures, such as mutual information. Furthermore, by providing closed form or computationally lighter algorithms, it may significantly reduce the computational burden of their implementation. We propose a modeling framework which aims to give new tools for the design of networks of new artificial self organizing, embodied and intelligent agents and the reverse engineering of natural ones. At this point, it represents much a theoretical conjecture and it has still to be experimentally verified whether this model will be useful in practice..