Can an AI model with no sense of embodiment learn choreographic instincts? Using motion capture of her own movements as training data, Mariel Pettee has worked with teams of collaborators to develop custom machine learning tools to generate new choreography that feels like her own. In this talk, she will discuss not only these models and their generative capabilities and pitfalls, but also the creative motivations and tensions that guided her while shaping this research trajectory. This is a story about engaging with AI as a creative collaborator. As an artist and researcher, Pettee tries to understand to what extent algorithms can reflect some of the most cherished parts of ourselves — and what to make of the reflection.

Mariel Pettee (Physics PhD, Yale University) is a Chamberlain Postdoctoral Research Fellow at Lawrence Berkeley National Laboratory and a visiting researcher at the Flatiron Institute Center for Computational Astrophysics in New York City. As a choreographer, director, dancer, and performer, she also uses ML as an artistic medium. As a dancer, she has performed in works by Kinetech Arts, Emily Coates, Renee Robinson, Brian Siebert, Elm City Dance Collective, the Bill T. Jones/Arnie Zane Company, Paul Taylor, Urban Bush Women, LeeSaar, and Jill Johnson. She was a choreographer-in-residence at Harvard University’s Dance Center in 2014. Prior to her PhD, she earned her Bachelors in Physics & Mathematics from Harvard University and her Masters in Physics at the University of Cambridge (Trinity College) as a Harvard-Cambridge Scholar.