Spatio-Temporal Graph Convolutional Networks: Optimised Temporal Architecture
This work addresses the optimization of temporal architectures in spatio-temporal graph networks, which is an incremental improvement for researchers in graph-based machine learning.
The paper tackles the problem of improving spatio-temporal graph convolutional networks by proposing a novel architecture that combines CNN and LSTM temporal blocks, and it provides an empirical comparison showing promising results across multiple datasets.
Spatio-Temporal graph convolutional networks were originally introduced with CNNs as temporal blocks for feature extraction. Since then LSTM temporal blocks have been proposed and shown to have promising results. We propose a novel architecture combining both CNN and LSTM temporal blocks and then provide an empirical comparison between our new and the pre-existing models. We provide theoretical arguments for the different temporal blocks and use a multitude of tests across different datasets to assess our hypotheses.