Time-o1: Time-Series Forecasting Needs Transformed Label Alignment
This addresses optimization issues in time-series forecasting for researchers and practitioners, but it is incremental as it builds on existing methods with a novel objective.
The paper tackled the challenges of label autocorrelation and excessive tasks in time-series forecasting by proposing Time-o1, a transformation-augmented learning objective that decorrelates label components and aligns significant ones, achieving state-of-the-art performance in experiments.
Training time-series forecast models presents unique challenges in designing effective learning objectives. Existing methods predominantly utilize the temporal mean squared error, which faces two critical challenges: (1) label autocorrelation, which leads to bias from the label sequence likelihood; (2) excessive amount of tasks, which increases with the forecast horizon and complicates optimization. To address these challenges, we propose Time-o1, a transformation-augmented learning objective tailored for time-series forecasting. The central idea is to transform the label sequence into decorrelated components with discriminated significance. Models are then trained to align the most significant components, thereby effectively mitigating label autocorrelation and reducing task amount. Extensive experiments demonstrate that Time-o1 achieves state-of-the-art performance and is compatible with various forecast models. Code is available at https://github.com/Master-PLC/Time-o1.