Speaker: Dr. Christian Walder (Data61, Australia)
Title: Neural Dynamic Programming for Musical Self Similarity
Abstract: We present a neural sequence model designed specifically for symbolic music. The model is based on a learned edit distance mechanism which generalises a classic recursion from computer science, leading to a neural dynamic program. Repeated motifs are detected by learning the transformations between them. We represent the arising computational dependencies using a novel data structure, the edit tree; this perspective suggests natural approximations which afford the scaling up of our otherwise cubic time algorithm.
|Date||October 10, 2018 (Wed) 14:00 - 15:30|