Identification of Invariant Sensorimotor Structures as a Prerequisite for the Discovery of Objects

TL;DR

This paper presents a computational model inspired by Sensorimotor Contingencies Theory that enables an agent to identify invariant structures in its sensorimotor experience, facilitating object perception through unsupervised learning and spectral clustering.

Contribution

It introduces a novel unsupervised, predictive approach to detect spatio-temporally invariant structures in sensorimotor data, advancing understanding of object perception mechanisms.

Findings

Invariant structures induce regularities in sensorimotor experience

Spectral clustering effectively captures these structures as connected subgraphs

The model demonstrates how an agent can learn object-like features without supervision

Abstract

Perceiving the surrounding environment in terms of objects is useful for any general purpose intelligent agent. In this paper, we investigate a fundamental mechanism making object perception possible, namely the identification of spatio-temporally invariant structures in the sensorimotor experience of an agent. We take inspiration from the Sensorimotor Contingencies Theory to define a computational model of this mechanism through a sensorimotor, unsupervised and predictive approach. Our model is based on processing the unsupervised interaction of an artificial agent with its environment. We show how spatio-temporally invariant structures in the environment induce regularities in the sensorimotor experience of an agent, and how this agent, while building a predictive model of its sensorimotor experience, can capture them as densely connected subgraphs in a graph of sensory states connected by motor commands. Our approach is focused on elementary mechanisms, and is illustrated with a set of simple experiments in which an agent interacts with an environment. We show how the agent can build an internal model of moving but spatio-temporally invariant structures by performing a Spectral Clustering of the graph modeling its overall sensorimotor experiences. We systematically examine properties of the model, shedding light more globally on the specificities of the paradigm with respect to methods based on the supervised processing of collections of static images.