TY - GEN
T1 - Neural dynamic programming for musical self similarity
AU - Walder, Christian J.
AU - Kim, Dongwoo
N1 - Publisher Copyright:
© 2018 by the Authors All rights reserved.
PY - 2018
Y1 - 2018
N2 - 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. We demonstrate our model on real and synthetic data; in all cases it out-performs a strong stacked long short-term memory benchmark.
AB - 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. We demonstrate our model on real and synthetic data; in all cases it out-performs a strong stacked long short-term memory benchmark.
UR - http://www.scopus.com/inward/record.url?scp=85057332536&partnerID=8YFLogxK
M3 - Conference contribution
T3 - 35th International Conference on Machine Learning, ICML 2018
SP - 8088
EP - 8096
BT - 35th International Conference on Machine Learning, ICML 2018
A2 - Krause, Andreas
A2 - Dy, Jennifer
PB - International Machine Learning Society (IMLS)
T2 - 35th International Conference on Machine Learning, ICML 2018
Y2 - 10 July 2018 through 15 July 2018
ER -