Explores graph neural networks, which have emerged as a powerful tool for machine learning on irregular structures, enabling novel applications across diverse fields within AI. Topics include essential neural network concepts, traditional neural networks' limitations when handling irregularly structured data, theoretical foundations, and practical applications of GNNs. Also covers advanced GNN architectures, generative models, dynamic graph networks, and transferability. Studies spectral-based and spatial-based GNN methods and evaluates models for various tasks through lectures, readings, assignments, and hands-on projects.