PT-Tuning: Bridging the Gap between Time Series Masked Reconstruction and Forecasting via Prompt Token Tuning

no code implementations • 7 Nov 2023 • Hao liu, Jinrui Gan, Xiaoxuan Fan, Yi Zhang, Chuanxian Luo, Jing Zhang, Guangxin Jiang, Yucheng Qian, Changwei Zhao, Huan Ma, Zhenyu Guo

In this paper, we first point out that the unification of task objectives and adaptation for task difficulty are critical for bridging the gap between time series masked reconstruction and forecasting.

Representation Learning Self-Supervised Learning +1