VMDNet: Time Series Forecasting with Leakage-Free Samplewise Variational Mode Decomposition and Multibranch Decoding
This work addresses issues in time series forecasting for energy domains, but it is incremental as it builds on existing decomposition techniques.
The paper tackled the problem of information leakage and hyperparameter tuning in time series forecasting using Variational Mode Decomposition (VMD), proposing VMDNet, which achieved state-of-the-art results on energy-related datasets, particularly when periodicity is strong.
In time series forecasting, capturing recurrent temporal patterns is essential; decomposition techniques make such structure explicit and thereby improve predictive performance. Variational Mode Decomposition (VMD) is a powerful signal-processing method for periodicity-aware decomposition and has seen growing adoption in recent years. However, existing studies often suffer from information leakage and rely on inappropriate hyperparameter tuning. To address these issues, we propose VMDNet, a causality-preserving framework that (i) applies sample-wise VMD to avoid leakage; (ii) represents each decomposed mode with frequency-aware embeddings and decodes it using parallel temporal convolutional networks (TCNs), ensuring mode independence and efficient learning; and (iii) introduces a bilevel, Stackelberg-inspired optimisation to adaptively select VMD's two core hyperparameters: the number of modes (K) and the bandwidth penalty (alpha). Experiments on two energy-related datasets demonstrate that VMDNet achieves state-of-the-art results when periodicity is strong, showing clear advantages in capturing structured periodic patterns while remaining robust under weak periodicity.