Meta-Forecasting by combining Global Deep Representations with Local Adaptation

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 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.

References in corpus (15)

• Adam: A Method for Stochastic Optimization
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
• Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting
• On First-Order Meta-Learning Algorithms
• Unsupervised Scalable Representation Learning for Multivariate Time Series
• Zoneout: Regularizing RNNs by Randomly Preserving Hidden Activations
• Differentiable Convex Optimization Layers
• Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?
• Deep Factors for Forecasting
• Forecasting: theory and practice
• Few-shot Learning for Time-series Forecasting
• Representation based meta-learning for few-shot spoken intent recognition
• Learning Feature Embeddings for Discriminant Model based Tracking