Skip to yearly menu bar Skip to main content


Virtual presentation / poster accept

A Time Series is Worth 64 Words: Long-term Forecasting with Transformers

Yuqi Nie · Nam Nguyen · Phanwadee Sinthong · Jayant Kalagnanam

Keywords: [ time series ] [ channel-independence ] [ forecasting ] [ transformer ] [ self-supervised learning ] [ representation learning ] [ Applications ]


Abstract:

We propose an efficient design of Transformer-based models for multivariate time series forecasting and self-supervised representation learning. It is based on two key components: (i) segmentation of time series into subseries-level patches which are served as input tokens to Transformer; (ii) channel-independence where each channel contains a single univariate time series that shares the same embedding and Transformer weights across all the series. Patching design naturally has three-fold benefit: local semantic information is retained in the embedding; computation and memory usage of the attention maps are quadratically reduced given the same look-back window; and the model can attend longer history. Our channel-independent patch time series Transformer (PatchTST) can improve the long-term forecasting accuracy significantly when compared with that of SOTA Transformer-based models. We also apply our model to self-supervised pre-training tasks and attain excellent fine-tuning performance, which outperforms supervised training on large datasets. Transferring of masked pre-training performed on one dataset to other datasets also produces SOTA forecasting accuracy.

Chat is not available.