Liu, Hongbin, and Lee, Ickjai (2020) Bridging the gap between training and inference for spatio-temporal forecasting. In: Frontiers in Artificial Intelligence and Applications (325) pp. 1316-1323. From: ECAI 2020: 24th European Conference on Artificial Intelligence, 29 August - 8 September 2020, Santiago, Spain.

Preview

PDF (Published Version) - Published Version Available under License Creative Commons Attribution Non-commercial. Download (515kB) | Preview

 

446

Abstract

Spatio-temporal sequence forecasting is one of the fundamental tasks in spatio-temporal data mining. It facilitates many real world applications such as precipitation now casting, city wide crowd flow prediction and air pollution forecasting. Recently, a few Seq2Seq based approaches have been proposed, but one of the drawbacks of Seq2Seq models is that, small errors can accumulate quickly along the generated sequence at the inference stage due to the different distributions of training and inference phase. That is because Seq2Seq models minimise single step errors only during training, however the entire sequence has to be generated during the inference phase which generates a discrepancy between training and inference. In this work, we propose a novel curriculum learning based strategy named Temporal Progressive Growing Sampling to effectively bridge the gap between training and inference for spatio-temporal sequence forecasting, by transformin the training process from a fully-supervised manner which utilises all available previous groundtruth values to a less-supervised manner which replaces some of theground-truth context with generated predictions. To do that we sam-ple the target sequence from midway outputs from intermediate models trained with bigger timescales through a carefully designed decaying strategy. Experimental results demonstrate that our proposed method better models long term dependencies and outperforms baseline approaches on two competitive datasets.

Actions (Repository Staff Only)

Item Control Page