Learning deep time-index models for time series forecasting

• Main Authors: WOO, Jiale Gerald, LIU, Chenghao, SAHOO, Doyen, KUMAR, Akshat, HOI, Steven
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2023
• Subjects: time-series; forecasting; deep learning; implicit neural representation; meta-learning; time-index; non-stationary; Artificial Intelligence and Robotics; Operations Research, Systems Engineering and Industrial Engineering
• Online Access: https://ink.library.smu.edu.sg/sis_research/8605; https://ink.library.smu.edu.sg/context/sis_research/article/9608/viewcontent/woo23b.pdf
• Institution: Singapore Management University

Deep learning has been actively applied to time series forecasting, leading to a deluge of new methods, belonging to the class of historicalvalue 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 timeindex 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 metaoptimization 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.