Tag
This paper proposes a hybrid WGAN-GA approach for refining generative graph topologies, using a genetic algorithm to correct residual structural deviations in GAN-based generated graphs, improving realism for synthetic graph synthesis and data augmentation.