Not All Data are Good Labels: On the Self-supervised Labeling for Time Series Forecasting
This work addresses the challenge of insufficient data exploitation in time series forecasting for domains like finance or weather, offering a novel method to enhance model generalization, though it appears incremental as it builds on existing self-supervised and regularization techniques.
The paper tackles the problem of time series forecasting models relying on high-quality data by proposing a self-supervised labeling approach that constructs candidate datasets and uses pseudo labels from reconstruction intermediates to improve generalization, with experiments on eleven real-world datasets showing consistent performance improvements across various backbone models.
Time Series Forecasting (TSF) is a crucial task in various domains, yet existing TSF models rely heavily on high-quality data and insufficiently exploit all available data. This paper explores a novel self-supervised approach to re-label time series datasets by inherently constructing candidate datasets. During the optimization of a simple reconstruction network, intermediates are used as pseudo labels in a self-supervised paradigm, improving generalization for any predictor. We introduce the Self-Correction with Adaptive Mask (SCAM), which discards overfitted components and selectively replaces them with pseudo labels generated from reconstructions. Additionally, we incorporate Spectral Norm Regularization (SNR) to further suppress overfitting from a loss landscape perspective. Our experiments on eleven real-world datasets demonstrate that SCAM consistently improves the performance of various backbone models. This work offers a new perspective on constructing datasets and enhancing the generalization of TSF models through self-supervised learning. The code is available at https://github.com/SuDIS-ZJU/SCAM.