Balancing Graph Embedding Smoothness in Self-Supervised Learning via Information-Theoretic Decomposition

Open in new window