Search papers, labs, and topics across Lattice.
This paper introduces the SMart framework for time series representation learning, which enhances existing methods by integrating a multi-phase recurrence plots recovery task and a source dataset selector for improved representation transfer. These innovations address limitations in previous approaches that relied on simpler self-supervised tasks and single-source datasets. Experimental results demonstrate that SMart significantly outperforms state-of-the-art models, achieving up to 19.5% reduction in mean absolute error for regression tasks and a 1.34% increase in classification accuracy.
SMart achieves a remarkable 19.5% reduction in regression error by leveraging multi-source datasets and innovative recovery tasks in time series representation learning.
Time series representation learning (TSRL) has attracted growing research interests in recent years. Two recent explorations in TSRL are: i) exploiting a transformer-based framework to learn time series; ii) instead of using only the targeted dataset, borrowing time series from other datasets to to facilitate representation transfer. While these two explorations are shown effective, the self-supervised time series recovery task in (i) and the single-source dataset used in (ii) are technically simple and thus can be enhanced with new ideas. In this work, we propose a new TSRL framework, namely multi-source multi-phase time series representation transfer (SMart), which has two novel mechanisms to address the aforementioned deficiencies: 1) a multi-phase recurrence plots recovery task, in three alternative modes, for guiding the encoder to embed time series dynamics into the time series representation; and 2) a source dataset selector to select multiple suitable source datasets to supplement the original target dataset for pre-training the TSRL encoder. Experimental results show that SMart outperforms several state-of-the-art models for time series representation learning, classification and regression on both uni-variate and multi-variate time series datasets, reducing mean absolute error up to 19.5% for time series regression, and increasing average accuracy up to 1.34\% for time series classification.
