Out-of-distribution Generalization for Learning Quantum Dynamics and Dynamical Simulation

Watch a conference talk from TQC 2023 exploring groundbreaking research on out-of-distribution generalization in Quantum Machine Learning (QML), where speaker Matthias C. Caro presents novel theoretical guarantees for training quantum neural networks to learn unknown unitaries using product state training data while generalizing to entangled states. Learn about a new QML-based algorithm for simulating quantum dynamics on near-term quantum hardware, supported by both rigorous theoretical analysis of resource requirements and experimental validation through classical simulations and quantum hardware implementations. Discover how these advances in generalization bounds and training methodologies are expanding the possibilities for quantum circuit compilation and NISQ-era quantum computing applications.