Making pictures with Artificial Intelligence | Helena Sarin | TEDxKonstanz

How does artificial intelligence sample images, store information, and re-assemble it in a single art piece? And what does this tell us about the way we perceive images and aesthetics? Helena Sarin is an artist who uses machine learning techniques - and more specifically GANs (Generative Adversarial Networks), as a new medium for picture making. Helena exploits the latent representations learned by generative models as a form of compressed memory, with its own unique aesthetic that can be both familiar and visually indeterminate.

Helena Sarin is a visual artist and software engineer. Although always involved with cutting edge technologies, Helena had continued doing commission work in watercolor and pastel and other types of “analog” art.. That is, until she discovered GANs. In addition to her work being exhibited at AI Art exhibitions across the world, she is a frequent speaker at machine learning conferences, for example for MIT, Library of Congress, Capitol One, Adobe Research and CMU.

Intro animation created by Mike Costelloe. Music intro from Phase 3 by Xylo-Ziko, used under CC BY-NC 4.0.

Video edited by Lilith Kugler.

Visual artist and software engineer, Helena Sarin has always been working with cutting edge technologies, first at Bell Labs, designing commercial communication systems, and for the last few years as an independent consultant, developing computer vision software using deep learning. While she has always worked in tech, Helena has been doing commission work in watercolor and pastel as well as in the applied arts like fashion, food and drink styling and photography. But art and software ran as parallel tracks in her life, all her art being analog… until she discovered GANs (Generative Adversarial Networks).

Since then generative models became her primary medium. She is a frequent speaker at ML/AI conferences, for the past year delivering invited talks at MIT, Library of Congress, Capitol One, Adobe Research. Her artwork was exhibited at AI Art exhibitions in Zurich, Dubai, Oxford, Shanghai and Miami, virtual AI gallery at Nvidia GTC gallery and was featured in number of publications including the recent issue of “Art In America” magazine. She published an artist book “The Book of GANesis” that was immediately sold out and now working on “The Book of veGAN” and “GANcommedia Erudita”. This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at https://www.ted.com/tedx