Unsupervised Transfer Learning for Time Series via Self-Predictive Modelling - First Results
Aswolinskiy W, Hammer B (2017)
In: Proceedings of the Workshop on New Challenges in Neural Computation (NC2). Machine Learning Reports, 03/2017. Bielefeld: Universität Bielefeld, CITEC.
Konferenzbeitrag
| Veröffentlicht | Englisch
Download
Autor*in
Einrichtung
Abstract / Bemerkung
Real-world machine learning applications must be able to adapt to systematic changes in the data, e.g. a new subject or sensor displacement. This can be seen as a form of transfer learning, where the goal is to reuse the old (source) model by adapting the new (target) data. This is a challenging task, if no labels for the target data are available. Here, we propose to use the structure of the source and target data to find a transformation from the source to target space in an unsupervised manner. Our preliminary experiments on multivariate time series data show the feasibility of the approach, but also its limits.
Stichworte
domain adaptation;
transductive transfer lerning;
time series classification;
predictive modelling;
echo state networks
Erscheinungsjahr
2017
Titel des Konferenzbandes
Proceedings of the Workshop on New Challenges in Neural Computation (NC2)
Serien- oder Zeitschriftentitel
Machine Learning Reports
Band
03/2017
Konferenz
Workshop on New Challenges in Neural Computation (NC2)
Konferenzort
Basel
Konferenzdatum
2017-09-12
ISSN
1865-3960
Page URI
https://pub.uni-bielefeld.de/record/2914141
Zitieren
Aswolinskiy W, Hammer B. Unsupervised Transfer Learning for Time Series via Self-Predictive Modelling - First Results. In: Proceedings of the Workshop on New Challenges in Neural Computation (NC2). Machine Learning Reports. Vol 03/2017. Bielefeld: Universität Bielefeld, CITEC; 2017.
Aswolinskiy, W., & Hammer, B. (2017). Unsupervised Transfer Learning for Time Series via Self-Predictive Modelling - First Results. Proceedings of the Workshop on New Challenges in Neural Computation (NC2), Machine Learning Reports, 03/2017 Bielefeld: Universität Bielefeld, CITEC.
Aswolinskiy, Witali, and Hammer, Barbara. 2017. “Unsupervised Transfer Learning for Time Series via Self-Predictive Modelling - First Results”. In Proceedings of the Workshop on New Challenges in Neural Computation (NC2). Vol. 03/2017. Machine Learning Reports. Bielefeld: Universität Bielefeld, CITEC.
Aswolinskiy, W., and Hammer, B. (2017). “Unsupervised Transfer Learning for Time Series via Self-Predictive Modelling - First Results” in Proceedings of the Workshop on New Challenges in Neural Computation (NC2) Machine Learning Reports, vol. 03/2017, (Bielefeld: Universität Bielefeld, CITEC).
Aswolinskiy, W., & Hammer, B., 2017. Unsupervised Transfer Learning for Time Series via Self-Predictive Modelling - First Results. In Proceedings of the Workshop on New Challenges in Neural Computation (NC2). Machine Learning Reports. no.03/2017 Bielefeld: Universität Bielefeld, CITEC.
W. Aswolinskiy and B. Hammer, “Unsupervised Transfer Learning for Time Series via Self-Predictive Modelling - First Results”, Proceedings of the Workshop on New Challenges in Neural Computation (NC2), Machine Learning Reports, vol. 03/2017, Bielefeld: Universität Bielefeld, CITEC, 2017.
Aswolinskiy, W., Hammer, B.: Unsupervised Transfer Learning for Time Series via Self-Predictive Modelling - First Results. Proceedings of the Workshop on New Challenges in Neural Computation (NC2). Machine Learning Reports. 03/2017, Universität Bielefeld, CITEC, Bielefeld (2017).
Aswolinskiy, Witali, and Hammer, Barbara. “Unsupervised Transfer Learning for Time Series via Self-Predictive Modelling - First Results”. Proceedings of the Workshop on New Challenges in Neural Computation (NC2). Bielefeld: Universität Bielefeld, CITEC, 2017.Vol. 03/2017. Machine Learning Reports.
Alle Dateien verfügbar unter der/den folgenden Lizenz(en):
Copyright Statement:
Dieses Objekt ist durch das Urheberrecht und/oder verwandte Schutzrechte geschützt. [...]
Volltext(e)
Access Level
Open Access
Zuletzt Hochgeladen
2019-09-06T09:18:52Z
MD5 Prüfsumme
42df26e0483500f53d4b4bd78a702760
Google Scholar