A decoder-only foundation model for time-series forecasting
This provides a general-purpose forecasting tool for domains like finance or climate, though it is incremental as it adapts existing NLP methods to time-series.
The paper tackles the problem of time-series forecasting by developing a decoder-only foundation model that achieves zero-shot performance close to state-of-the-art supervised models on various public datasets.
Motivated by recent advances in large language models for Natural Language Processing (NLP), we design a time-series foundation model for forecasting whose out-of-the-box zero-shot performance on a variety of public datasets comes close to the accuracy of state-of-the-art supervised forecasting models for each individual dataset. Our model is based on pretraining a patched-decoder style attention model on a large time-series corpus, and can work well across different forecasting history lengths, prediction lengths and temporal granularities.