Circuit expressivity alone isn't sufficient for quantum training. Discrete CNOT routing controls which optimization directions are accessible, causing gradient obstructions that can be diagnosed and repaired through same-point augmentation.

#QuantumAlgorithms #VariationalQuantum #Research

NOVQS framework shows multiple shallow parameterized quantum circuits match or exceed single deep circuit performance for quantum dynamics. Reveals depth-circuit count trade-off enabling quantum advantage on near-term hardware.

#QuantumChemistry #VariationalQuantum #Research

Novel variational framework combining deep learning with Monte Carlo achieves 0.45% energy accuracy for the 2D Hubbard model and reveals charge-spin stripe patterns in doped regimes using efficient low-rank matrix shifts.

#QuantumSimulation #VariationalQuantum #Research

Novel method for quantum simulation that tracks selected observables rather than full quantum state, extending accurate simulation time by up to 4.2× using same measurement budget. Avoids ancillary qubits for implementation.

#QuantumSimulation #VariationalQuantum #Research