SimMTM: A Simple Pre-Training Framework for Masked Time-Series Modeling

TL;DR

SimMTM introduces a novel pre-training framework for masked time-series modeling that leverages manifold learning to improve reconstruction and downstream task performance.

Contribution

It proposes a simple, effective method that relates masked modeling to manifold learning, enhancing the recovery of temporal variations in time series.

Findings

Achieves state-of-the-art results in forecasting and classification tasks.

Improves representation learning by recovering masked points through neighbor aggregation.

Effective in both in-domain and cross-domain settings.

Abstract

Time series analysis is widely used in extensive areas. Recently, to reduce labeling expenses and benefit various tasks, self-supervised pre-training has attracted immense interest. One mainstream paradigm is masked modeling, which successfully pre-trains deep models by learning to reconstruct the masked content based on the unmasked part. However, since the semantic information of time series is mainly contained in temporal variations, the standard way of randomly masking a portion of time points will seriously ruin vital temporal variations of time series, making the reconstruction task too difficult to guide representation learning. We thus present SimMTM, a Simple pre-training framework for Masked Time-series Modeling. By relating masked modeling to manifold learning, SimMTM proposes to recover masked time points by the weighted aggregation of multiple neighbors outside the manifold, which eases the reconstruction task by assembling ruined but complementary temporal variations from multiple masked series. SimMTM further learns to uncover the local structure of the manifold, which is helpful for masked modeling. Experimentally, SimMTM achieves state-of-the-art fine-tuning performance compared to the most advanced time series pre-training methods in two canonical time series analysis tasks: forecasting and classification, covering both in- and cross-domain settings.