Label-efficient Time Series Representation Learning: A Review
This is an incremental survey that organizes existing methods for time series analysis, aiding researchers in understanding and advancing the field.
The paper reviews label-efficient time series representation learning, addressing the problem of limited labeled data by categorizing existing approaches as in-domain or cross-domain based on external data reliance, and it surveys recent advances, limitations, and future directions without presenting new experimental results.
Label-efficient time series representation learning, which aims to learn effective representations with limited labeled data, is crucial for deploying deep learning models in real-world applications. To address the scarcity of labeled time series data, various strategies, e.g., transfer learning, self-supervised learning, and semi-supervised learning, have been developed. In this survey, we introduce a novel taxonomy for the first time, categorizing existing approaches as in-domain or cross-domain, based on their reliance on external data sources or not. Furthermore, we present a review of the recent advances in each strategy, conclude the limitations of current methodologies, and suggest future research directions that promise further improvements in the field.