Pixel by Pixel: The Art of CHANCE

In our introductory editors’ letter, we highlighted the new directions we wanted to explore. One of those new directions is to highlight the intersection between data science and other domains. We begin this exploration with a generative art challenge.

“A person paints with their brains and not with their hands.” This quote, attributed to Michelangelo (and modified by us), captures the spirit of our art challenge—Pixel by Pixel: The Art of CHANCE. For this challenge, we want our readers to use your brains and your computational skills to create generative art.

The CHANCE masthead includes this mission: “using data to advance science, education, and society.” We are introducing this generative art challenge as a next step in meeting this mission. Although the definition of data science continues to evolve, it is not debatable that the successful data science workflow requires computational skills and knowledge. Our goal for Pixel by Pixel: The Art of CHANCE is to encourage and showcase these skills and knowledge while highlighting the creative applications of this discipline.

As a report from the National Academies of Science and Medicine stated: “Data science is emerging as a field that is revolutionizing science and industries alike. Work across nearly all domains is becoming more data-driven, affecting both the jobs that are available and the skills that are required. As more data and ways of analyzing [it] become[s] available, more aspects of the economy, society, and daily life will become dependent on data.” The generative art challenge provides the opportunity to showcase these essential skills and have fun doing it.

Artistic Inspiration

The theme for this challenge is creativity, so we are purposefully establishing very few rules or constraints. However, we would like you to identify an artist to serve as your generative art muse. The artists Vincent van Gogh, Pablo Picasso, Jackson Pollock, Edvard Munch, Salvador Dali, and Corita Kent are the group from which to select your muse. These brief characteristics are what influenced the decision to include them as motivating artists.

  • • van Gogh was recognized for his use of color and bold technique.
  • • Picasso is credited with co-inventing collage and Cubism; both of these styles are recognized as challenging the traditional view of art.
  • • Pollock’s inclusion emphasizes that creativity often involves risk-taking; succinctly put, “The risks and the creative approaches he took, led future artists to create with passion, as opposed to trying to follow set boundaries or guidelines.”
  • • Munch said, “Nature is not only all that is visible to the eye…it also includes the inner pictures of the soul.” He was a well-known Symbolist painter likely to inspire the emotion and ideas reflected in your generative art.
  • • A Dali Paintings essay states, “…Dali will always stand out as one of the very few twentieth-century painters who combines profound respect for the traditions of the past with intensely modern feelings.” The same should be true of the work submitted for this challenge.
  • • In her book Learning by Heart: Teachings to Free the Creative Spirit, pop artist Kent offers rules for students and teachers, including “Consider everything an experiment,” which seems to be excellent advice for this generative art challenge.

This quote from the gallery label for an untitled artwork describes the process used by early generative artists: “Chance and Control: Art in the Age of Computers (2018) GEORG NEES (1926–2016). Untitled, Published by Werkstatt Edition, Kroll, Germany, 1970. This screenprint was produced from a unique plotter drawing. The plotter was operated by feeding punched tape into a computer that used the instructions to direct a pen across a drawing surface. As the computer had no screen, Nees would not have been able to fully anticipate the appearance Screenprint after a computer-generated artwork programmed on a Siemens System 4004 and drawn on a Zuse-Graphomat flatbed drawing machine, 1970.”

What is Generative Art?

According to art historian Jason Bailey, generative art—also known as computer or algorithmic art—is “art programmed using a computer that intentionally introduces randomness as part of its creation process.” He addressed two issues sometimes used to label it as not being considered as “art,” the main one being that the art created using computer code is deterministic, with the artist having absolute control over the execution.

However, as statisticians, we know how to introduce randomness into algorithms, so we can achieve and explore different results. As Bailey said, “The truth is that generative artists skillfully control both the magnitude and the locations of randomness introduced into the artwork.”

The idea of using algorithms and computers to generate works of art has been around since the 1960s, starting with the generative art created by Georg Nees, Frieder Nake, and A. Michael Noll (sometimes known as the 3Ns of digital art; http://dada.compart-bremen.de for historical information on these artists and their tools). As you might expect, they did not have the digital tools and hardware (e.g., high-resolution monitors and tablets) for creating and sharing computer art that we have today. Instead, they used a computer to direct printers to physically draw their art on paper, which did not give them the opportunity to try to explore many different outcomes in search of their perfect creations.

Software Tools for Creating Generative Art

Today, we have better computational tools to create and share algorithmic art. There are two main computer languages for data science and statistics: R and Python. For online content and the latest tools and references associated with the Art of CHANCE, go to https://bit.ly/chance-contest.

Katharina Brunner developed an R package called generativeart that is available for download on github (GitHub – cutterkom/generativeart: Create Generative Art with R). For those who want to use an interactive approach with R Shiny, see Martin Pedersen’s GenerativeR package (GitHub – MartinMSPedersen/GenerativeR: Generative Art using R+Shiny). Koen Derks developed an R package posted to CRAN titled aRtsy: Generative Art with ggplot2.

We could not find any specific Python packages for generative art as there are for R. However, some websites provide examples of creating algorithmic art. One is called Ice or Fire, where the author provides Python code to create random flow fields shown as flowing and colorful lines. A blog post by Jonathan Chaffer uses a Python version of turtle graphics to create some generative art that might serve as inspiration.

Image created by Donna LaLonde using GenerativeR.


We want to encourage participation, so we have minimal requirements. To enter the contest, provide the image you have created, your code, and any other information (e.g., seed for random generator used) the judges would need to reproduce the image.

Because this is a generative art contest, the creation of your image must involve some aspect of randomness. You are free to use the tools we have described in this article and in the supplemental material—or to write your own.

The submission categories are: middle and high school students, undergraduate students, graduate students, and professionals. Awards will be given in each category. Each submission should include a short descriptive essay of the creative process, including which artist served as the inspiration and motivation for the tools used. Submissions are due on or before August 31, 2023. A submission form will is available here. Send questions about the contest to the editors at chancemag.editor@gmail.com.

Further Reading

AB 101: Historical Figures in Generative Art—Georg Nees.

Bailey, Jason. 2018. Why Love Generative Art? Artnome.

Chaffer, Jonathan. 2021. Atomic Object, Learn How to Make Generative Art: From Zero to random(). atomicobject.com.

Chua, Jessica. 2021. Do Machines Dream of Creativity? A Beginner’s Guide to Generative Art. The Artling.



Ice or Fire. 2021. Generative Art with Python.

Kent, C., and Steward, J. 1992. Consider everything an experiment. New York, NY: Bantam Books.

About the Authors

Wendy Martinez is the senior mathematical statistician for data science in the Research and Methodology Directorate of the U.S. Census Bureau. Previously, she served as director of the Mathematical Statistics Research Center in the Bureau of Labor Statistics (BLS) and worked in several research positions in the Department of Defense. Her research interests include computational statistics, exploratory data analysis, text analysis, and data visualization. She received her PhD in computational sciences and informatics with an emphasis on computational statistics at George Mason University in 1995. She completed a master of science degree in engineering at George Washington University in 1991 and a bachelor of science degree in mathematics and physics at Cameron University in 1989.

Donna LaLonde is the associate executive director at the ASA, where she works with colleagues to advance the vision and mission of the association. Before joining the ASA, she was a faculty member at Washburn University and served in various administrative positions, including interim chair of the Education Department and associate vice president for academic affairs. At the ASA, LaLonde has supported activities associated with presidential initiatives, professional development, and accreditation.

Back to Top

Tagged as: , , , ,

Leave a Response

Please note: comment moderation is enabled and may delay your comment. There is no need to resubmit your comment.