Hybrid Variational Autoencoder for Time Series Forecasting
This work addresses a specific limitation in time series forecasting for applications requiring both local and dynamic modeling, representing an incremental improvement over existing VAE-based methods.
The paper tackled the problem of jointly learning local patterns and temporal dynamics in time series forecasting, proposing a hybrid variational autoencoder (HyVAE) that achieved better forecasting results than various counterpart methods and its own variants on four real-world datasets.
Variational autoencoders (VAE) are powerful generative models that learn the latent representations of input data as random variables. Recent studies show that VAE can flexibly learn the complex temporal dynamics of time series and achieve more promising forecasting results than deterministic models. However, a major limitation of existing works is that they fail to jointly learn the local patterns (e.g., seasonality and trend) and temporal dynamics of time series for forecasting. Accordingly, we propose a novel hybrid variational autoencoder (HyVAE) to integrate the learning of local patterns and temporal dynamics by variational inference for time series forecasting. Experimental results on four real-world datasets show that the proposed HyVAE achieves better forecasting results than various counterpart methods, as well as two HyVAE variants that only learn the local patterns or temporal dynamics of time series, respectively.