musicking deep reinforcement learning

Scurto, 2019—2022
reflexive essay, music, computer science

A research-creation with musicking and deep reinforcement learning. It builds on the scientific study of two AI models, an audio VAE and the Co-Explorer, and on musical experiments led in the frame of the ægo performance. I discuss how deep reinforcement learning can be seen as a form of sonic comprovisational agent, enabling to compose sound spaces and improvise through feedback. I then reflect on how this opened my performer’s expectations away from instrumental control of sound, to deepen my listening of sound, and learn spiritual unification with music.

Year
2019—2022

Credits
The project was developed with Axel Chemla—Romeu-Santos in collaboration with the ISMM and ACIDS groups of IRCAM, in the context of a PhD thesis at Sorbonne Université.

Publications
Chapter at LNCS Springer (2021)
Paper at TENOR (2022)

This slideshow requires JavaScript.