Robust quantum optimizer with full connectivity.

Area of Science:

• Quantum Computing
• Optimization Algorithms
• Condensed Matter Physics

Background:

• Quantum annealing offers potential speedups for hard optimization problems compared to classical methods.
• Current superconducting qubit quantum annealers face challenges with limited qubit connectivity and decoherence.
• These limitations restrict the types of optimization problems that can be directly implemented and solved.

Purpose of the Study:

• To propose a new quantum annealer architecture that addresses connectivity and decoherence issues.
• To demonstrate a robust method for encoding qubits in continuous variable degrees of freedom.
• To enable the direct implementation of a wider range of complex optimization problems.

Main Methods:

• Encoding qubits in continuous variable degrees of freedom.
• Leveraging flux quantization to achieve all-to-all qubit connectivity.
• Simulating the solution of a number partitioning problem on the proposed architecture.

Main Results:

• The proposed architecture achieves all-to-all connectivity with tunable capabilities.
• The continuous variable encoding demonstrates robustness against decoherence and dissipation.
• A small instance of a fully connected, NP-hard number partitioning problem was successfully simulated.

Conclusions:

• The developed architecture offers a promising solution to overcome key limitations in current quantum annealers.
• This approach enhances the feasibility of applying quantum annealing to a broader class of difficult optimization problems.
• The robustness demonstrated suggests potential for practical applications in solving complex computational challenges.