Journal
20TH IEEE INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS (ICMLA 2021)
Volume -, Issue -, Pages 422-428Publisher
IEEE
DOI: 10.1109/ICMLA52953.2021.00072
Keywords
Semi-supervised Learning; Time Series Classification; Data Augmentation
Funding
- Bavarian Ministry of Economic Affairs, Regional Development and Energy through the Center for Analytics -Data -Applications (ADA-Center) [20-3410-29-8]
- German Federal Ministry of Education and Research (BMBF) [01IS18036A]
Ask authors/readers for more resources
This study investigates the feasibility of transferring state-of-the-art deep semi-supervised models from image to time series classification, emphasizing necessary model adaptations and tailored data augmentation strategies. Through evaluations on large public time series classification problems, the transferred semi-supervised models show significant performance gains, especially in scenarios with very few labeled samples.
While deep semi-supervised learning has gained much attention in computer vision, limited research exists on its applicability in the time series domain. In this work, we investigate the transferability of state-of-the-art deep semi-supervised models from image to time series classification. We discuss the necessary model adaptations, in particular an appropriate model backbone architecture and the use of tailored data augmentation strategies. Based on these adaptations, we explore the potential of deep semi-supervised learning in the context of time series classification by evaluating our methods on large public time series classification problems with varying amounts of labeled samples. We perform extensive comparisons under a decidedly realistic and appropriate evaluation scheme with a unified reimplementation of all algorithms considered, which is yet lacking in the field. We find that these transferred semi-supervised models show significant performance gains over strong supervised, semi-supervised and self-supervised alternatives, especially for scenarios with very few labeled samples.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available