What is it about?
Creativity, and its apparent inexplicability, has always fascinated the human kind. Several theories and approaches have been developed during the last century to define and evaluate it. In this article, we propose to use deep learning to quantify the creativity of human or artificial products. In particular, three different techniques are developed to measure three of the fundamental criteria of creativity: value, novelty, and surprise. Preliminary studies on American poetry suggest that deep learning may be suitable for such a problem.
Featured Image
Photo by Patrick Tomasso on Unsplash
Why is it important?
It is one of the first attempts of defining an automatic way to measure creativity. In addition, it is based on broadly accepted theories of human creativity, i.e., that value, novelty and surprise are central for the creative products.
Perspectives
Studying creativity is such an amazing activity, especially from an AI perspective. It makes you really understand that the ability to generate human-level artifacts is not enough, and there are many other aspects to consider. I hope this articles makes a valuable contribution for this multidisciplinary and timeless field.
Giorgio Franceschelli
Universita degli Studi di Bologna
Read the Original
This page is a summary of: DeepCreativity: measuring creativity with deep learning techniques, Intelligenza Artificiale, December 2022, IOS Press,
DOI: 10.3233/ia-220136.
You can read the full text:
Contributors
The following have contributed to this page
