Can I read DAG-GCN Directed Acyclic Causal Graph Discovery From Real World Data Using Graph Convolutional Networks - 2023 on EtoBox?

DAG-GCN Directed Acyclic Causal Graph Discovery From Real World Data Using Graph Convolutional Networks - 2023 by Ping-Feng Xu is a document available to read on EtoBox.

What is DAG-GCN Directed Acyclic Causal Graph Discovery From Real World Data Using Graph Convolutional Networks - 2023 about?

The paper presents a novel model called DAG-GCN for causal discovery using Graph Convolutional Networks (GCN) to address challenges in real-world data. It improves upon existing methods by incorporating techniques like He-Initialization and batch normalization to mitigate issues with vanishing adjacency matrices. Experimental results on the Sachs dataset demonstrate that the proposed model outperforms previous approaches in terms of Structural Hamming Distance (SHD) and True Positive Rate (TPR).

Author

Ping-Feng Xu

Language

EN