Transfer learning for hierarchical forecasting: Reducing computational efforts of M5 winning method
dc.contributor.author | Wellens, Arnoud P. | |
dc.contributor.author | Udenio, Maxi | |
dc.contributor.author | Boute, Robert | |
dc.date.accessioned | 2021-10-05T13:40:42Z | |
dc.date.available | 2021-10-05T13:40:42Z | |
dc.date.issued | 2022 | en_US |
dc.identifier.issn | 0169-2070 | |
dc.identifier.doi | 10.1016/j.ijforecast.2021.09.011 | |
dc.identifier.uri | http://hdl.handle.net/20.500.12127/6974 | |
dc.description.abstract | The winning machine learning methods of the M5 Accuracy competition demonstrated high levels of forecast accuracy compared to the top-performing benchmarks in the history of the Mcompetitions. Yet, large-scale adoption is hampered due to the signi cant computational requirements to model, tune, and train these state-of-the-art algorithms. To overcome this major issue, we discuss the potential of transfer learning (TL) to reduce the computational e ort in hierarchical forecasting and provide proof of concept that TL can be applied on M5 top-performing methods. We demonstrate our easy-to-use TL framework on the recursive store level LightGBM models of the M5 winning method and attain similar levels of forecast accuracy with roughly 25% less training time. Our ndings provide evidence for a novel application of TL to facilitate practical applicability of the M5 winning methods in large-scale settings with hierarchically structured data. | en_US |
dc.language.iso | en | en_US |
dc.publisher | Elsevier | en_US |
dc.subject | M5 Accuracy Competition | en_US |
dc.subject | Computational Requirements | en_US |
dc.subject | Transfer Learning | en_US |
dc.subject | LightGBM | en_US |
dc.subject | Hierarchical Forecasting | en_US |
dc.title | Transfer learning for hierarchical forecasting: Reducing computational efforts of M5 winning method | en_US |
dc.identifier.journal | International Journal of Forecasting | en_US |
dc.source.volume | 38 | |
dc.source.issue | 4 | |
dc.source.beginpage | 1482 | |
dc.source.endpage | 1491 | |
dc.contributor.department | KU Leuven, Research Center for Operations Management, Naamsestraat 69, Leuven, 3000, Belgium | en_US |
vlerick.knowledgedomain | Operations & Supply Chain Management | en_US |
vlerick.typearticle | Journal article with impact factor | en_US |
vlerick.vlerickdepartment | TOM | en_US |
dc.identifier.vperid | 102358 | en_US |