DOBRO : a prediction error correcting robot under drifts

A. Maslov, H.T. Lam, M. Pechenizkiy, E. Bouillet, T. Kärkkäinen

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

We propose DOBRO, a light online learning module, which is equipped with a smart correction policy helping making decision to correct or not the given prediction depending on how likely the correction will lead to a better prediction performance. DOBRO is a standalone module requiring nothing more than a time series of prediction errors and it is flexible to be integrated into any black-box model to improve its performance under drifts. We performed evaluation in a real-world application with bus arrival time prediction problem. The obtained results show that DOBRO improved prediction performance significantly meanwhile it did not hurt the accuracy when drift does not happen
Original languageEnglish
Title of host publicationSAC '16 Proceedings of the 31st Annual ACM Symposium on Applied Computing, 4-8 April 2016, Pisa, Italy
Place of PublicationNew York
PublisherAssociation for Computing Machinery, Inc
Pages945-948
Number of pages4
ISBN (Print)978-1-4503-3739-7
DOIs
Publication statusPublished - 2016

Fingerprint Dive into the research topics of 'DOBRO : a prediction error correcting robot under drifts'. Together they form a unique fingerprint.

  • Cite this

    Maslov, A., Lam, H. T., Pechenizkiy, M., Bouillet, E., & Kärkkäinen, T. (2016). DOBRO : a prediction error correcting robot under drifts. In SAC '16 Proceedings of the 31st Annual ACM Symposium on Applied Computing, 4-8 April 2016, Pisa, Italy (pp. 945-948). New York: Association for Computing Machinery, Inc. https://doi.org/10.1145/2851613.2851888