Grounding object perception in a naive agent's sensorimotor experience

TL;DR

This paper proposes a method for grounding object perception in a naive agent's sensorimotor experience, showing how objects can emerge as consistent sensorimotor networks without prior knowledge.

Contribution

It introduces an algorithm that enables a naive agent to discover objects as sensorimotor patterns purely through exploration, without predefined models.

Findings

Objects emerge as stable sensorimotor networks during exploration

The approach does not rely on prior knowledge or feature extraction

The algorithm demonstrates the development of object perception from sensorimotor experience

Abstract

Artificial object perception usually relies on a priori defined models and feature extraction algorithms. We study how the concept of object can be grounded in the sensorimotor experience of a naive agent. Without any knowledge about itself or the world it is immersed in, the agent explores its sensorimotor space and identifies objects as consistent networks of sensorimotor transitions, independent from their context. A fundamental drive for prediction is assumed to explain the emergence of such networks from a developmental standpoint. An algorithm is proposed and tested to illustrate the approach.