Reliable localized on-line learning in non-stationary environments
Autor(en): | Buschermöhle, A. Brockmann, W. |
Herausgeber: | Angelov, P. Filev, D. Kasabov, N. Lughofer, E. Klement, E.P. Saminger-Platz, S. Iglesias, J.A. Sayed-Mouchaweh, M. |
Stichwörter: | Electric power transmission networks; Intelligent systems, Global input; Hyper-parameter; Input-output relations; Local training; Non-stationary environment; Online learning; Power grids; State-of-the-art algorithms, E-learning | Erscheinungsdatum: | 2014 | Herausgeber: | Institute of Electrical and Electronics Engineers Inc. | Journal: | 2014 IEEE Conference on Evolving and Adaptive Intelligent Systems, EAIS 2014 - Conference Proceedings | Zusammenfassung: | On-line learning allows to adapt to changing non-stationary environments. But typically with on-line learning a hypothesis of the data relation is adapted based on a stream of single local training examples, continuously changing the global input-output relation. Hence with these single examples the whole hypothesis is revised incrementally, which might be harmful to the overall predictive quality of the learned model. Nevertheless, for a reliable adaptation, the learned model must yield good predictions in every step. There for, the IRMA approach to online learning enables an adaptation that reliably incorporates a new example with a stringent local, but minimal global influence on the input-output relation. The main contribution of this paper is twofold. First, it presents an extension of IRMA regarding the setup of the stiffness, i.e. its hyper-parameter. Second, the IRMA approach is investigated for the first time on a non-trivial real-world application in a non-stationary environment. It is compared with state of the art algorithms on predicting future electric loads in a power grid where a continuous adaptation is necessary to adapt to season and weather conditions. The results show that the performance is increased significantly by IRMA. © 2014 IEEE. |
Beschreibung: | Conference of 2014 IEEE Conference on Evolving and Adaptive Intelligent Systems, EAIS 2014 ; Conference Date: 2 June 2014 Through 4 June 2014; Conference Code:109713 |
ISBN: | 9781479933471 | DOI: | 10.1109/eais.2014.6867475 | Externe URL: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-84920545316&doi=10.1109%2feais.2014.6867475&partnerID=40&md5=b2758c759f449d69a3157fdfe630ac19 |
Zur Langanzeige
Seitenaufrufe
3
Letzte Woche
0
0
Letzter Monat
2
2
geprüft am 01.06.2024