Meta-Forecasting by combining Global Deep Representations with Local Adaptation

Riccardo Grazzi, Valentin Flunkert, David Salinas, Tim Januschowski, Matthias Seeger, Cedric Archambeau

Keywords: time-series, meta-learning, closed-form, solvers

Abstract: 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 of the 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.

One-sentence Summary: Combining an RNN representation with a close-form adaptation allows to compute accurate forecasts for out of sample time-series