Szczegóły publikacji
Opis bibliograficzny
LIFEWATCH: lifelong wasserstein change point detection / Kamil FABER, Roberto Corizzo, Bartłomiej ŚNIEŻYŃSKI, Michael Baron, Nathalie Japkowicz // W: IJCNN 2022 [Dokument elektroniczny] : International Joint Conference on Neural Networks : Padua, Italy, 18–23 July 2022 : proceedings / IEEE. — Wersja do Windows. — Dane tekstowe. — Piscataway : IEEE, cop. 2022. — (Proceedings of ... International Joint Conference on Neural Networks ; ISSN 2161-4393). — Konferencja zorganizowana w ramach IEEE World Congress on Computational Intelligence (IEEE WCCI 2022). — e-ISBN: 978-1-7281-8671-9. — s. 1-8. — Bibliogr., Abstr.
Autorzy (5)
- AGHFaber Kamil
- Corizzo Roberto
- AGHŚnieżyński Bartłomiej
- Baron Michael
- Japkowicz Nathalie
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 142986 |
|---|---|
| Data dodania do BaDAP | 2023-01-02 |
| DOI | 10.1109/IJCNN55064.2022.9892891 |
| Rok publikacji | 2022 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Wydawca | Institute of Electrical and Electronics Engineers (IEEE) |
| Konferencja | IEEE International Joint Conference on Neural Networks 2022 |
| Czasopismo/seria | Proceedings of ... International Joint Conference on Neural Networks |
Abstract
Change point detection methods offer a crucial ca-pability in modern data analysis tasks characterized by evolving time series data in the form of data streams. Recent interest in lifelong learning showed the importance of acquiring knowledge and identifying new occurring tasks in a continually evolving environment. Although this setting could benefit from a timely identification of changes, existing change point detection methods are unable to recognize recurring tasks, which is a necessary condition in lifelong learning. In this paper, we attempt to fill this gap by proposing LIFEWATCH, a novel Wasserstein-based change point detection approach with memory capable of modeling multiple data distributions in a fully unsupervised manner. Our method does not only detect changes, but discriminates between changes characterized by the appearance of a new task and changes that rather describe a recurring or previously seen task. An extensive experimental evaluation involving a large number of benchmark datasets shows that LIFEWATCH outperforms state-of-the-art methods for change detection while exploiting the characterization of detected changes to correctly identify tasks occurring in complex scenarios characterized by recurrence in lifelong consolidation settings.
