This course introduces machine learning for art and creativity. It is a broad survey of the tools, techniques, and theory needed to understand emerging AI technology and re-appropriate it for critical inquiry and creative exploration. The contents include an accessible introduction to how modern neural networks function and their real-time and non-real-time applications, as well as an overview of current state-of-the-art techniques in deep learning. We\’ll build interactive systems which incorporate real-time learning into creative code environments such as Processing, p5.js, openFrameworks, Max/MSP, and PureData, as well as control software instruments which produce music and visual art. We will also explore the frontiers of generative models such as GANs and autoencoders, showing how these methods can learn how to synthesize complex and information-rich images, sounds, and text. Course materials will be based on the tools and instructional guides being developed on ml4a.github.io, along with a suite of deep learning libraries that perform important and novel tasks. A high-level, non-comprehensive introduction to coding machine learning in Python using modern deep learning frameworks like PyTorch and Tensorflow will be included. An introduction and overview of RunwayML will also be part of the course. Students will be provided with all of the code and supporting materials. Although this course has no official prerequisites, students will find it useful to catch up on fundamental computer science skills, including using a terminal and coding basic Python. One or more optional sessions for students who wish to catch up on or refresh these skills will be offered within the first two weeks.

ITPG-GT.2994.1 (22604) | Instructor: Gene Kogan | Tues 3:20pm to 6:15pm | Meeting Pattern: 12 | Start Date