The Construction of Reality in an AI: A Review

TL;DR

This paper reviews constructivist AI inspired by Piaget, highlighting recent progress in knowledge representation and learning algorithms aimed at enabling lifelong learning in agents.

Contribution

It summarizes prior work, recent developments, and introduces a new framework for constructive AI that processes sensory input into semantic and episodic memory.

Findings

Progress in knowledge representations and learning algorithms

Development of a semantic memory network linked to episodic data

Encouragement for further research in lifelong learning AI

Abstract

AI constructivism as inspired by Jean Piaget, described and surveyed by Frank Guerin, and representatively implemented by Gary Drescher seeks to create algorithms and knowledge structures that enable agents to acquire, maintain, and apply a deep understanding of the environment through sensorimotor interactions. This paper aims to increase awareness of constructivist AI implementations to encourage greater progress toward enabling lifelong learning by machines. It builds on Guerin's 2008 "Learning Like a Baby: A Survey of AI approaches." After briefly recapitulating that survey, it summarizes subsequent progress by the Guerin referents, numerous works not covered by Guerin (or found in other surveys), and relevant efforts in related areas. The focus is on knowledge representations and learning algorithms that have been used in practice viewed through lenses of Piaget's schemas, adaptation processes, and staged development. The paper concludes with a preview of a simple framework for constructive AI being developed by the author that parses concepts from sensory input and stores them in a semantic memory network linked to episodic data. Extensive references are provided.