Loading…
This event has ended. Visit the official site or create your own event on Sched.
View analytic
Monday, June 4 • 1:30pm - 3:00pm
Paper 2.3

Log in to save this to your schedule and see who's attending!

Exploring Continuous Time Recurrent Neural Networks through Novelty Search
by Steffan Carlos Ianigro & Oliver Bown


Within this paper, we aim to expand on prior research into the use of Continuous Time Recurrent Neural Networks (CTRNNs) as evolvable generators of musical structures such as audio waveforms. This type of neural network has a compact structure and is capable of producing a large range of temporal dynamics. Due to these properties, we believe that CTRNNs could provide a genotype structure for an EA that offers musicians many creative possibilities for the exploration of sound. In prior work, we have explored the use of interactive and target-based EA designs to tap into the creative possibilities of CTRNNs. Our results have shown promise for the use of CTRNNs in the audio domain. However, we feel neither evolutionary algorithm (EA) designs allow both open-ended discovery and effective navigation of the CTRNN audio search space by musicians. Within this paper, we explore the possibility of using novelty search as an alternative algorithm that facilitates both open-ended and rapid discovery of the CTRNN creative search space.

Speakers
avatar for Oliver Bown

Oliver Bown

Senior Lecturer, UNSW Faculty of Art & Design, Interactive Media Lab
I am a researcher and maker working with creative technologies. I come from a highly diverse academic background spanning social anthropology, evolutionary and adaptive systems, music informatics and interaction design, with a parallel career in electronic music and digital art spanning... Read More →


Attendees (6)