Learning Deep Time-index Models for Time Series Forecasting
This addresses the challenge of accurate and efficient time series forecasting for applications in domains like finance or weather prediction, representing an incremental improvement by adapting time-index models with deep learning.
The paper tackles the problem of time series forecasting by proposing DeepTime, a meta-optimization framework for learning deep time-index models, which overcomes limitations of naive deep time-index models and achieves competitive results with state-of-the-art methods while being highly efficient.
Deep learning has been actively applied to time series forecasting, leading to a deluge of new methods, belonging to the class of historical-value models. Yet, despite the attractive properties of time-index models, such as being able to model the continuous nature of underlying time series dynamics, little attention has been given to them. Indeed, while naive deep time-index models are far more expressive than the manually predefined function representations of classical time-index models, they are inadequate for forecasting, being unable to generalize to unseen time steps due to the lack of inductive bias. In this paper, we propose DeepTime, a meta-optimization framework to learn deep time-index models which overcome these limitations, yielding an efficient and accurate forecasting model. Extensive experiments on real world datasets in the long sequence time-series forecasting setting demonstrate that our approach achieves competitive results with state-of-the-art methods, and is highly efficient. Code is available at https://github.com/salesforce/DeepTime.