Spatio-temporal time series forecasting has attracted great attentions in various fields, including climate, power, and traffic forecasting. Recently, Spatio-temporal Graph Neural Networks (STGNNs) ...

It is important to note that the graph G is entirely characterized by its adjacency matrix, thereby establishing a significant correlation with the transition matrix of the corresponding Markov chain.

However, most of the existing graph neural network methods based on static adjacency matrices are difficult to capture individual differences and remote brain region interactions, thus limiting their ...