Print Email Facebook Twitter Transfer Learning of Artist Group Factors to Musical Genre Classification Title Transfer Learning of Artist Group Factors to Musical Genre Classification Author Kim, Jaehun (TU Delft Multimedia Computing) Won, Minz (Pompeu Fabra University) Serra, Xavier (Pompeu Fabra University) Liem, C.C.S. (TU Delft Multimedia Computing) Date 2018 Abstract The automated recognition of music genres from audio information is a challenging problem, as genre labels are subjective and noisy. Artist labels are less subjective and less noisy, while certain artists may relate more strongly to certain genres. At the same time, at prediction time, it is not guaranteed that artist labels are available for a given audio segment. Therefore, in this work, we propose to apply the transfer learning framework, learning artist-related information which will be used at inference time for genre classification. We consider different types of artist-related information, expressed through artist group factors, which will allow for more efficient learning and stronger robustness to potential label noise. Furthermore, we investigate how to achieve the highest validation accuracy on the given FMA dataset, by experimenting with various kinds of transfer methods, including single-task transfer, multi-task transfer and finally multi-task learning. Subject music information retrievalmulti-task learningtransfer learningneural network To reference this document use: http://resolver.tudelft.nl/uuid:a48bba52-d5d2-4f5d-a84f-cd8884fcc1e4 DOI https://doi.org/10.1145/3184558.3191823 Publisher International World Wide Web Conferences Steering Committee, Republic and Canton of Geneva, Switzerland ISBN 978-1-4503-5640-4 Source WWW'18 Companion Proceedings of the The Web Conference 2018 Event WWW 2018, 2018-04-23 → 2018-04-27, Lyon, France Part of collection Institutional Repository Document type conference paper Rights © 2018 Jaehun Kim, Minz Won, Xavier Serra, C.C.S. Liem Files PDF 45678209u_p1929_kim.pdf 1.55 MB Close viewer /islandora/object/uuid:a48bba52-d5d2-4f5d-a84f-cd8884fcc1e4/datastream/OBJ/view