GASP: a genetic algorithm for state preparation on quantum computers

Area of Science:

• Quantum Computing
• Quantum Information Science
• Algorithm Development

Background:

• Efficient quantum state preparation is crucial for quantum algorithms, especially in the NISQ era with limited quantum resources.
• Low-depth quantum circuits are essential for implementation on current noisy intermediate-scale quantum (NISQ) devices.
• Existing state preparation methods face challenges in resource efficiency and circuit depth.

Purpose of the Study:

• To introduce a novel genetic algorithm for state preparation (GASP) to generate efficient, low-depth quantum circuits.
• To enable the initialization of quantum computers into specified quantum states with high fidelity.
• To compare GASP's performance against established state initialization techniques.

Main Methods:

• Utilizing a genetic algorithm with a basis set of single-qubit rotations and CNOT gates.
• Systematically generating quantum circuits to synthesize target states to a required fidelity.
• Implementing and comparing GASP with IBM Qiskit's exact synthesis method on simulated and physical quantum devices.

Main Results:

• GASP generates quantum circuits with significantly lower depth and gate counts compared to other methods for a given accuracy.
• The algorithm demonstrates superior performance in synthesizing various quantum states, including Gaussian and W-states.
• Results show a consistent reduction in the number of gates required for accurate state preparation.

Conclusions:

• GASP offers a more efficient approach to quantum state preparation, crucial for advancing NISQ algorithms.
• The method's ability to produce lower-depth circuits mitigates error accumulation, enhancing implementation fidelity.
• GASP provides a competitive and effective alternative to existing state initialization techniques in quantum computing.