"Emergence" by Lia Coleman

Lia Coleman is a Chinese-American artist and researcher based in Pittsburgh, Pennsylvania. Coleman’s work centers on the interplay of AI technology, art & design, coding, and ethics. They actively conduct research on creative AI at Carnegie Mellon’s Robotics Institute and publish guides for artists to responsibly use machine learning.

Coleman’s work explores themes of collective memory and loss. In assembling datasets, they attempt to record and archive the past — a process which is inherently never complete. They embrace the unpredictability of neural networks, as a reflection of their own experience of grappling with the elusive nature of truth and the past. Coleman is an alum of the Massachusetts Institute of Technology (MIT) and the School For Poetic Computation. They have also served as an adjunct professor at the Rhode Island School of Design (RISD). Their work has been shown internationally in Dubai, Germany, Malta, Saudi Arabia, Canada, and the United States, and has been featured by Vox, Wired, Tribeca Film Festival, Mozilla Festival, Science Gallery Detroit, New York University, the NeurIPS Conference, and Gray Area. Their writing on AI art has been published by Princeton Architectural Press, DISEÑA, and Neocha Magazine.

In “Emergence”, Lia Coleman uses a neural network to expand upon their musings on time, the collective, and remembrance, as it relates to their Chinese-American heritage. Through mesmerizing visual transformations that mimic the biological emergence of life, Coleman invites viewers to reflect upon the powerful interconnectedness that weaves together our collective history on this planet.

Through use of Python-based tools such as PyTorch, ffmpeg, and moviepy, Coleman creates the artwork in a process where time plays a vital role. In an act of digital archival of the current era, Coleman collected and curated a dataset of hundreds of images of Chinese papercuts. In the training loop of a machine learning system, where past generations inextricably influence the current generation, each artwork emerges from a different moment in time. While each edition represents a different snapshot during the evolutionary process of creation, the full collection is of the same lineage, of the same family tree.

Embracing the concept of the collective, Coleman pays homage to the profound influence of those who came before us, such as our ancestors and artists of previous generations, while also highlighting the symbiotic relationships we cultivate with contemporaries like friends, family, and collaborators. The undulating black & white aesthetic of “Emergence" finds analogues in the work of Ryoichi Kurokawa and Ryoji Ikeda.

Each collected edition of “Emergence” includes a set of bonus files from the artist, consisting of 5 checkpoints of Coleman's complete machine learning model at different stages of training in the form of .pkl files, allowing collectors to generate their own images and animations in the spirit of collective co-creation.