Formal Verification of Neural Networks: Potential and Advantage with Quantum Computing

Artificial intelligence (AI) has seen widespread adoption across sectors, including manufacturing, finance, and healthcare. However, deploying AI in safety-critical applications raises concerns due to their vulnerability to input perturbations. The thesis not only contributes to the theoretical foundations of formal verification and quantum-enhanced methods but also provides practical insights into their implementation and potential impact on the broader field of AI safety and security.