Informing Artificial Intelligence Generative Techniques using Cognitive Theories of Human Creativity

TL;DR

This paper explores integrating cognitive theories of human creativity into AI generative models, demonstrating how concepts like honing theory and intrinsic motivation can enhance computational art, and vice versa.

Contribution

It presents a novel synthesis of deep learning and cognitive theories, showing how human creativity models can improve AI-generated art and inform psychological understanding.

Findings

Implementation of honing theory and intrinsic motivation in generative art

Impact of cognitive concepts on the quality of AI-generated outputs

Cross-fertilization ideas between AI creativity and psychology

Abstract

The common view that our creativity is what makes us uniquely human suggests that incorporating research on human creativity into generative deep learning techniques might be a fruitful avenue for making their outputs more compelling and human-like. Using an original synthesis of Deep Dream-based convolutional neural networks and cognitive based computational art rendering systems, we show how honing theory, intrinsic motivation, and the notion of a 'seed incident' can be implemented computationally, and demonstrate their impact on the resulting generative art. Conversely, we discuss how explorations in deep learn-ing convolutional neural net generative systems can inform our understanding of human creativity. We conclude with ideas for further cross-fertilization between AI based computational creativity and psychology of creativity.