Search range in experimental quantum annealing

Abstract

We construct an Ising Hamiltonian with an engineered energy landscape such that it has a local energy minimum which is near to the true global minimum solution, and further away from a false minimum. Using a technique established in previous experiments, we design our experiment such that (at least on timescales relevant to our study) the false minimum is reached preferentially in forward annealing due to high levels of quantum fluctuations. This allows us to demonstrate the key principle of reverse annealing, that the solution space can be searched locally, preferentially finding nearby solutions, even in the presence of a false minimum. The techniques used here are, to the best of our knowledge, distinct from previously used experimental techniques, and allow us to probe the fundamental search range of the device in a way which has not been previously explored. We perform these experiments on two flux qubit quantum annealers, one with higher noise levels than the other. We find evidence that the lower noise device is more likely to find the more distant energy minimum (the false minimum in this case), suggesting that reducing noise fundamentally increases the range over which flux qubit quantum annealers are able to search. Our work explains why reducing the noise leads to improved performance on these quantum annealers. This supports the idea that these devices may be able to search over broad regions of the solution space quickly, one of the core reasons why quantum annealers are viewed as a potential avenue for a quantum computational advantage.

arXiv comment13 pages, 10 figures, now matching the final PRA submission, associated data repository can be found at http://doi.org/10.15128/r2kp78gg42f

References in corpus (25)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• Array Programming with NumPy
• Probing many-body dynamics on a 51-atom quantum simulator
• Observation of topological phenomena in a programmable lattice of 1,800 qubits
• Experimental signature of programmable quantum annealing
• Reverse Quantum Annealing Approach to Portfolio Optimization Problems
• Computational Role of Multiqubit Tunneling in a Quantum Annealer
• Entanglement in a quantum annealing processor
• Scaling advantage in quantum simulation of geometrically frustrated magnets
• Glassy Chimeras could be blind to quantum speedup: Designing better benchmarks for quantum annealing machines
• Prospects for Quantum Enhancement with Diabatic Quantum Annealing
• Quantum Annealing Applied to De-Conflicting Optimal Trajectories for Air Traffic Management
• Quantum Annealing Implementation of Job-Shop Scheduling
• Modernizing Quantum Annealing using Local Searches
• How "Quantum" is the D-Wave Machine?
• Probing Noise in Flux Qubits via Macroscopic Resonant Tunneling
• Reverse Annealing for Nonnegative/Binary Matrix Factorization
• Quantum annealing with longitudinal bias fields
• Maximum-Entropy Inference with a Programmable Annealer
• Thermalization, freeze-out and noise: deciphering experimental quantum annealers
• An energetic perspective on rapid quenches in quantum annealing
• Two quantum Ising algorithms for the Shortest Vector Problem: one for now and one for later
• Low Rank Non-Negative Matrix Factorization with D-Wave 2000Q
• Search range in experimental quantum annealing
• Quantum-Assisted Genetic Algorithm

Cited by in corpus (15)

• Hybrid quantum-classical algorithms in the noisy intermediate-scale quantum era and beyond
• Understanding domain-wall encoding theoretically and experimentally
• Standard quantum annealing outperforms adiabatic reverse annealing with decoherence
• Search range in experimental quantum annealing
• Fluctuation guided search in quantum annealing
• Cyclic Quantum Annealing: Searching for Deep Low-Energy States in 5000-Qubit Spin Glass
• Comparing the hardness of MAX 2-SAT problem instances for quantum and classical algorithms
• Performance benefits of increased qubit connectivity in quantum annealing 3-dimensional spin glasses
• Grover Speedup from Many Forms of the Zeno Effect
• Load Balancing For High Performance Computing Using Quantum Annealing
• Quantum annealing sampling with a bias field
• Learning-Driven Annealing with Adaptive Hamiltonian Modification for Solving Large-Scale Problems on Quantum Devices
• The Perturbed Ferromagnetic Chain: A Tuneable Test of Quantum Hardness in the Transverse-Field Ising Model
• Noise Effects on Diabatic Quantum Annealing Protocols
• Quantum annealing and condensed matter physics