21M.369 / 21M.569, Fall 2024
Instructor: Anna Huang
Meets: TR 11:00-12:30, 24-033F
Explores algorithms and interactions needed for designing music systems that can support human-AI partnerships in music making. Surveys techniques in generative modeling, learning from human feedback, reinforcement learning (RL), and social RL. Also looks at interactive systems and visualizations to distill abstractions that empower human-AI collaboration as well as creative team work. Experience with Python and/or Javascript and basic machine learning paradigms recommended. Culminates in a final team project. Students taking graduate version complete different assignments. Enrollment limited to 12.
About the instructor
Anna Huang is Robert N. Noyce Career Development Professor MIT Music and Theater Arts and Electrical Engineering-Computer Science. Previously, she spent eight years with Magenta at Google Brain and Google DeepMind, spearheading efforts in generative modeling, reinforcement learning and human-computer interaction to support human-AI partnerships in music-making. She is the creator of Music Transformer and Coconet (which powered the Bach Doodle), and also a judge for the AI Song Contest.