Deep Time Series Forecasting: Tools and Open Challenges

Deep learning has recently shown promise for time series forecasting, raising questions about when it truly outperforms traditional statistical methods. In this session, recent advances in deep time series models will be reviewed, the open-source libraries TSLib and OpenLTM will be introduced, and future research directions in deep learning-based time series forecasting will be discussed.

Published

June 27, 2025

Deep Time Series Forecasting: Tools and Open Challenges

June 27, 11:30 AM

Can neural networks outperform traditional statistical methods, or do they merely add complexity without consistent gains? In this talk, I will review the progress in deep time series models in the past few years and introduce TSLib and OpenLTM, which are widely used open-source libraries for deep learning researchers. Moreover, this talk will also cover some opening discussions for future research on deep learning-based time series forecasting.