Computational Creativity: A Survey of Methods in Machine Learning Applied to the Arts

Date

2022-05-25

Journal Title

Journal ISSN

Volume Title

Publisher

Yeshiva University

YU Faculty Profile

Abstract

The arts and computer science are traditionally seen as two separate disciplines, unlikely to collide. Those who participate in art and music are thought of as right-brained, due to the creativity involved in these activities. Computer science, on the other hand, is considered a left-brained endeavor because of the intense analytical thinking it requires. Personally, I am both an artist and a computer scientist. I thought it would be an interesting and worthwhile undertaking to explore how the arts and computer science interplay.¶ In this survey, I introduce different machine learning methods that are applied to the fields of art and music. to frame the discussion, we introduce neural networks which are the foundation for machine learning algorithms. Then we proceed to a specific kind of neural network called the Convolutional Neural Network. This is essential for understanding this research, as many machine learning algorithms for the arts utilize Convolutional Neural Networks. After that, we present two applications of machine learning to art, including style transfer and art identification. In terms of machine learning and music, we present a different type of neural network called the Long Short Term Memory Network (LSTM). Lastly, we explore both music generation and music genre identification using machine learning algorithms. (Introduction)

Description

Undergraduate honors thesis / YU Only

Keywords

computer science, arts and computers

Citation

Inslicht, N. (2022, May 25). Computational Creativity: A Survey of Methods in Machine Learning Applied to the Arts. Undergraduate honors thesis, Yeshiva University.