Forecaster: A Graph Transformer for Forecasting Spatial and Time-Dependent Data
Forecaster: A Graph Transformer for Forecasting Spatial and Time-Dependent Data
复制标题
DOI：
10.3233/faia200231
复制
发表时间：
2019-09
期刊：
影响因子：
--
通讯作者：
Y. Li;J. Moura
中科院分区：
文献类型：
--
作者：
Y. Li;J. Moura
空间和时间相关的数据在许多应用中是令人感兴趣的。由于其复杂的空间依赖性、远程时间依赖性、数据非平稳性和数据异构性，这项任务很困难。为了解决这些挑战，我们提出了预测器，一个图形Transformer架构。具体来说，我们从学习图的结构开始，该结构简单地表示不同位置的数据之间的空间依赖性。基于图的拓扑结构，我们稀疏的Transformer占的空间依赖性，长距离的时间依赖性，数据的非平稳性，和数据的异构性的强度。我们在预测出租车乘车需求的问题中评估了Forecaster，并表明我们提出的架构显着优于最先进的基线。
Spatial and time-dependent data is of interest in many applications. This task is difficult due to its complex spatial dependency, long-range temporal dependency, data non-stationarity, and data heterogeneity. To address these challenges, we propose Forecaster, a graph Transformer architecture. Specifically, we start by learning the structure of the graph that parsimoniously represents the spatial dependency between the data at different locations. Based on the topology of the graph, we sparsify the Transformer to account for the strength of spatial dependency, long-range temporal dependency, data non-stationarity, and data heterogeneity. We evaluate Forecaster in the problem of forecasting taxi ride-hailing demand and show that our proposed architecture significantly outperforms the state-of-the-art baselines.