Diversify: A General Framework for Time Series Out-of-Distribution Detection and Generalization
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2024-06 · 46 (6) · 4534-4550
DOI: 10.1109/tpami.2024.3355212 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024-06. Verfasst von Wang Lu und 6 weiteren Autor:innen. Die Arbeit wurde bisher 44-mal zitiert (Crossref) und nennt 95 Literaturangaben. Gefördert von: National Key Research and Development Plan of China, Beijing Municipal Science & Technology Commission, Chinese Academy of Sciences.
Abstract (Originalsprache)
Time series remains one of the most challenging modalities in machine learning research. Out-of-distribution (OOD) detection and generalization on time series often face difficulties due to their non-stationary nature, wherein the distribution changes over time. The dynamic distributions within time series present significant challenges for existing algorithms, especially in identifying invariant distributions, as most focus on scenarios where domain information is provided as prior knowledge. This paper aims to address the issues induced by non-stationarity in time series through the exploration of subdomains within a complete dataset for generalized representation learning. We propose Diversify, a general framework, for OOD detection and generalization on dynamic distributions of time series. Diversify operates through an iterative process: first identifying the ’worst-case’ latent distribution scenario, then working to minimize the gaps between these latent distributions. We implement Diversify by combining existing OOD detection methods according to either extracted features or outputs of models for detection while we also directly utilize outputs for classification. Theoretical insights support the framework's validity. Extensive experiments are conducted on seven datasets with different OOD settings across gesture recognition, speech commands recognition, wearable stress and affect detection, and sensor-based human activity recognition. Qualitative and quantitative results demonstrate that Diversify learns more generalized features and significantly outperforms other baselines.
Förderung
- National Key Research and Development Plan of China
- Beijing Municipal Science & Technology Commission
- Chinese Academy of Sciences
- Colleges and Universities in Shandong Province
Zitieren
APA
Lu, W., Wang, J., Sun, X., Chen, Y., Ji, X., Yang, Q., & Xie, X. (2024). Diversify: A General Framework for Time Series Out-of-Distribution Detection and Generalization. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(6), 4534-4550. https://doi.org/10.1109/tpami.2024.3355212
Vancouver
Lu W, Wang J, Sun X, Chen Y, Ji X, Yang Q, et al. Diversify: A General Framework for Time Series Out-of-Distribution Detection and Generalization. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024;46(6):4534-4550. doi:10.1109/tpami.2024.3355212
BibTeX
@article{Lu2024,
author = {Lu, Wang and Wang, Jindong and Sun, Xinwei and Chen, Yiqiang and Ji, Xiangyang and Yang, Qiang and Xie, Xing},
title = {Diversify: A General Framework for Time Series Out-of-Distribution Detection and Generalization},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2024},
volume = {46},
number = {6},
pages = {4534--4550},
doi = {10.1109/tpami.2024.3355212}
}
EUVIMED