Towards Lightweight Time Series Forecasting: a Patch-wise Transformer with Weak Data Enriching

TL;DR

LiPFormer is a lightweight, efficient Transformer model for time series forecasting that incorporates easy-to-access context information to improve accuracy and is suitable for edge deployment.

Contribution

The paper introduces LiPFormer, a simplified, lightweight Transformer with a novel attention mechanism and a weak data enriching module that enhances forecasting accuracy without high complexity.

Findings

Outperforms state-of-the-art methods in accuracy on nine datasets

Reduces parameter count, training time, and GPU memory usage

Enables fast inference suitable for edge devices

Abstract

Patch-wise Transformer based time series forecasting achieves superior accuracy. However, this superiority relies heavily on intricate model design with massive parameters, rendering both training and inference expensive, thus preventing their deployments on edge devices with limited resources and low latency requirements. In addition, existing methods often work in an autoregressive manner, which take into account only historical values, but ignore valuable, easy-to-obtain context information, such as weather forecasts, date and time of day. To contend with the two limitations, we propose LiPFormer, a novel Lightweight Patch-wise Transformer with weak data enriching. First, to simplify the Transformer backbone, LiPFormer employs a novel lightweight cross-patch attention and a linear transformation-based attention to eliminate Layer Normalization and Feed Forward Network, two heavy components in existing Transformers. Second, we propose a lightweight, weak data enriching module to provide additional, valuable weak supervision to the training. It enhances forecasting accuracy without significantly increasing model complexity as it does not involve expensive, human-labeling but using easily accessible context information. This facilitates the weak data enriching to plug-and-play on existing models. Extensive experiments on nine benchmark time series datasets demonstrate that LiPFormer outperforms state-of-the-art methods in accuracy, while significantly reducing parameter scale, training duration, and GPU memory usage. Deployment on an edge device reveals that LiPFormer takes only 1/3 inference time compared to classic Transformers. In addition, we demonstrate that the weak data enriching can integrate seamlessly into various Transformer based models to enhance their accuracy, suggesting its generality.