AI Triumph: 80% Accuracy Achieved with MNIST and GNN!

This is a fantastic achievement! The project successfully combined formal methods with machine learning to classify MNIST data using Graph Neural Networks (GNNs), achieving an impressive 80% accuracy. The methodical approach, detailed in the article, offers valuable insights for other researchers.

• •Successfully used Graph Neural Networks (GNNs) for MNIST classification.
• •Achieved 80.01% accuracy.
• •Employs a structured development process combining KAOS, Alloy, and Python.