Date of Award

2019

Document Type

Thesis

Degree Name

Bachelors

Department

Natural Sciences

First Advisor

Doucette, John

Area of Concentration

Computer Science

Abstract

This thesis focuses on the intersection of Artificial Intelligence (AI) and creativity . It investigates how AI can be used to model and understand creativity. The investigation starts with a literature review of historical and modern perspectives on creativity, and concludes that the varied definitions and perceptions of creativity rely on subjective value judgements. Therefore, a biologically rooted model of creativity should be based on the systems which underlie the emergence of subjective value judgements like that of novelty perception. A survey of the use of AI in art-making follows, showing that traditional computationalist and machine learning methods are not biologically plausible accounts of creativity. Instead, an embodied approach, using the tools of dynamical systems (DS) theory, is pursued. This thesis shows how analyzing an agent as a dynamical system gives insight into the emergence of unique subjective value judgements, such as novelty and aesthetic judgements, to stimulus. In order to explore the potential for the use of DS theory in creative AI, a GoPiGo3 wheeled robot programmed to determine motion with equations of chaotic and oscillatory systems is used to make visual art. This thesis concludes that the embodied perspective and the tools of dynamical systems provide a rich framework for understanding and engineering the emergence of the subjective value judgements which underlie creativity.

Download

Share

COinS