Coherence-Based Label Propagation over Time Series for Accelerated Active Learning

Abstract

Time-series data are ubiquitous these days, but lack of the labels in time-series data is regarded as a hurdle for its broad applicability. Meanwhile, active learning has been successfully adopted to reduce the labeling efforts in various tasks. Thus, this paper addresses an important issue, time-series active learning. Inspired by the temporal coherence in time-series data, where consecutive data points tend to have the same label, our label propagation framework, called TCLP, automatically assigns a queried label to the data points within an accurately estimated time-series segment, thereby significantly boosting the impact of an individual query. Compared with traditional time-series active learning, TCLP is shown to improve the classification accuracy by up to 7.1 times when only 0.8% of data points in the entire time series are queried for their labels.

Cite

Text

Shin et al. "Coherence-Based Label Propagation over Time Series for Accelerated Active Learning." International Conference on Learning Representations, 2022.

Markdown

[Shin et al. "Coherence-Based Label Propagation over Time Series for Accelerated Active Learning." International Conference on Learning Representations, 2022.](https://mlanthology.org/iclr/2022/shin2022iclr-coherencebased/)

BibTeX

@inproceedings{shin2022iclr-coherencebased,
  title     = {{Coherence-Based Label Propagation over Time Series for Accelerated Active Learning}},
  author    = {Shin, Yooju and Yoon, Susik and Kim, Sundong and Song, Hwanjun and Lee, Jae-Gil and Lee, Byung Suk},
  booktitle = {International Conference on Learning Representations},
  year      = {2022},
  url       = {https://mlanthology.org/iclr/2022/shin2022iclr-coherencebased/}
}