record

Thesis Info

LABS ID
00936
Thesis Title
Modes of Expression in Artworks Involving Computational Processes with an Emphasis on Neural Networks
Author
Hye Min Cho
2nd Author
3rd Author
Degree
MFA
Year
2019
Number of Pages
57
University
UCLA
Thesis Supervisor
Casey Reas
Supervisor e-mail
Other Supervisor(s)
Chandler McWilliams, Jennifer Steinkamp
Language(s) of Thesis
English
Department / Discipline
Design | Media Art
Copyright Ownership
Languages Familiar to Author
Chinese, English, Korean
URL where full thesis can be found
hyeminchochocom.files.wordpress.com/2019/07/thesis-print-07-hye-min-cho.pdf
Keywords
Abstract: 200-500 words
There are four main parts in this paper. In “Human Perception and Machine-mediated Perception,” links between human perception and machine-mediated perception are explored. It starts with a brief tour of human perception and its relationship to the human body, then explores the quality of associations possible through machine learning, followed by comparing and contrasting machine associations with processes of the human mind. In combination of that analysis, I clarify what we can learn from how our perception as active and mobile animals differ from that of a machine, and that machine-mediated perception can also act as a mirror that allows us to peer into the digitally integrated beings that we are starting to become. “Surrealist Methods of Creation and their Connection to Machine Association” explores parallels between machine based operations and surrealist methods of creation. It starts with a short introduction of Surrealism, Surrealists methods and Freud. The rest of the chapter analyzes the similarity and differences of surrealist methods and machine based operations, posing the question, “what can be gained from considering machine based operations as an extension of human abilities and thoughts?” “Traditional East Asian Painting and Image Classification Neural Networks” is devoted to forming parallels between traditional East Asian paintings and image classification neural networks. Scattered perspective is introduced as an example of a technique that shares similarities with the image processing methods of many common neural networks. The last main part focuses on the author’s work 石 (Rock), 财 (Wealth), 壶 (Kettle), 笔 (Writing brush), 松 (Pine). This chapter is a documentation on the methods used to generate the work. Both conceptual and technical aspects of the work are documented in a loosely chronological order.