Don't overfit the history -- Recursive time series data augmentation
This addresses generalization issues in time series analysis for practitioners, though it is incremental as it builds on existing augmentation concepts.
The paper tackles overfitting in time series learning by introducing the Recursive Interpolation Method (RIM), a data augmentation framework that generates new samples while preserving original dynamics, achieving strong performance on real-world tasks like regression and classification.
Time series observations can be seen as realizations of an underlying dynamical system governed by rules that we typically do not know. For time series learning tasks, we need to understand that we fit our model on available data, which is a unique realized history. Training on a single realization often induces severe overfitting lacking generalization. To address this issue, we introduce a general recursive framework for time series augmentation, which we call Recursive Interpolation Method, denoted as RIM. New samples are generated using a recursive interpolation function of all previous values in such a way that the enhanced samples preserve the original inherent time series dynamics. We perform theoretical analysis to characterize the proposed RIM and to guarantee its test performance. We apply RIM to diverse real world time series cases to achieve strong performance over non-augmented data on regression, classification, and reinforcement learning tasks.