Zero-shot and few-shot time series forecasting with ordinal regression recurrent neural networks
This addresses the challenge of data scarcity in time series forecasting for domains with limited observations, though it appears incremental as it builds on existing RNN frameworks.
The paper tackles the problem of time series forecasting with limited data by proposing a novel RNN-based model that learns a shared feature embedding over quantised time series, enabling accurate and reliable forecasting in zero-shot and few-shot scenarios.
Recurrent neural networks (RNNs) are state-of-the-art in several sequential learning tasks, but they often require considerable amounts of data to generalise well. For many time series forecasting (TSF) tasks, only a few dozens of observations may be available at training time, which restricts use of this class of models. We propose a novel RNN-based model that directly addresses this problem by learning a shared feature embedding over the space of many quantised time series. We show how this enables our RNN framework to accurately and reliably forecast unseen time series, even when there is little to no training data available.