Tag: Causal Graph Neural Networks
Browse our exclusive articles!
SCGNN: Enhancing Graph Neural Networks with Granular-ball Computing
Discover SCGNN, a novel graph neural network method using granular-ball computing to boost semantic consistency and reduce noise in graph learning.
GCGNet: Advanced Time Series Forecasting with Exogenous Data
Discover GCGNet, a novel model enhancing time series forecasting accuracy by integrating temporal and channel correlations with exogenous variables.
Graph Rewiring Techniques to Fix GNN Over-Squashing
Explore graph rewiring methods to overcome over-squashing and over-smoothing in GNNs, enhancing information flow and model performance.
Decoupled Relation Alignment for Heterogeneous Graph Models
Boost heterogeneous graph model performance with Decoupled Relation Subspace Alignment (DRSA) for better cross-domain feature and relation handling.
TypeBandit: Efficient Attribute Completion in Heterogeneous GNNs
Discover TypeBandit, a novel method for effective attribute completion in heterogeneous graph neural networks with type-level context allocation.