Meta-Forecasting by combining Global Deep Representations with Local Adaptation
This addresses the limitation of deep learning methods in forecasting new, unseen time series, which is crucial for real-world applications where classical methods are more applicable.
The paper tackles the problem of poor out-of-sample forecasting accuracy in deep learning methods for time series by proposing Meta-GLAR, a meta-learning approach that combines global deep representations with local adaptation, achieving competitive state-of-the-art accuracy in empirical evaluations.
While classical time series forecasting considers individual time series in isolation, recent advances based on deep learning showed that jointly learning from a large pool of related time series can boost the forecasting accuracy. However, the accuracy of these methods suffers greatly when modeling out-of-sample time series, significantly limiting their applicability compared to classical forecasting methods. To bridge this gap, we adopt a meta-learning view of the time series forecasting problem. We introduce a novel forecasting method, called Meta Global-Local Auto-Regression (Meta-GLAR), that adapts to each time series by learning in closed-form the mapping from the representations produced by a recurrent neural network (RNN) to one-step-ahead forecasts. Crucially, the parameters ofthe RNN are learned across multiple time series by backpropagating through the closed-form adaptation mechanism. In our extensive empirical evaluation we show that our method is competitive with the state-of-the-art in out-of-sample forecasting accuracy reported in earlier work.