Phase transitions in a programmable quantum spin glass simulator

Harris, Richard E.; Sato, Yuki; Berkley, A. J.; Reis, Maurício Sedrez dos; Altomare, Fabio; Amin, M. H. S.; Boothby, Kelly; Bunyk, P.; Deng, Chunqing; Enderud, C.; Huang, Shuiyuan; Hoskinson, Emile; Johnson, Mark W.; Ladizinsky, E.; Ladizinsky, N.; Lanting, T.; Li, R.; Medina, T.; Molavi, Reza; Neufeld, R. B.; Oh, T.; Павлов, И. В.; Perminov, I.; Poulin-Lamarre, Gabriel; Rich, Chris; Smirnov, Anatoly Yu.; Swenson, L. J.; Tsai, Nicholas; Volkmann, Mark H.; Whittaker, Jed D.; Yao, Jing-Shing

doi:10.1126/science.aat2025

Cited by 258 publications

(221 citation statements)

References 44 publications

Supporting

Mentioning

213

Contrasting

Unclassified

Order By: Relevance

“…With the proper rotation of the basis, we have solved the master equations and theoretically confirmed the main experimental findings of [33]. The results of the paper may be useful for understanding a dissipative evolution of various systems, from chromophores in quantum biology [30][31][32] to qubits in real-world quantum processors [14,15,44].…”

Section: Discussionsupporting

confidence: 54%

“…We are interested in dissipative evolution of a quantum annealer [8,14,15,33,37] treated as a system of N qubits coupled to a heat bath. The qubits are described by the Hamiltonian:…”

Section: The Hamiltonianmentioning

confidence: 99%

“…At the end of the evolution, the system will occupy a low-energy state of the final Hamiltonian, which may represent a solution to an optimization or a sampling problem. Another important direction of research is related to the simulation of large quantum systems with various quantum hardware [10] including quantum dots [11], ion traps [12,13], and lattices of coupled superconducting qubits [14,15].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Theory of open quantum dynamics with hybrid noise

Smirnov

Amin

2018

New J. Phys.

Self Cite

View full text Add to dashboard Cite

We develop a theory to describe dynamics of a non-stationary open quantum system interacting with a hybrid environment, which includes high-frequency and low-frequency noise components. One part of the system-bath interaction is treated in a perturbative manner, whereas the other part is considered exactly. This approach allows us to derive a set of master equations where the relaxation rates are expressed as convolutions of the Bloch-Redfield and Marcus formulas. Our theory enables analysis of systems that have extremely small energy gaps in the presence of a realistic environment. As an illustration, we apply the theory to the 16 qubit quantum annealing problem with dangling qubits (Dickson et al 2013 Nat. Commun. 4 1903) and show qualitative agreement with experimental results.[28, 29] is not applicable. We provide an intuitively appealing and computationally convenient form for the transition rates in terms of a convolution between Redfield and Marcus formulas. As an example, we investigate a dissipative evolution of a 16 qubit system strongly interacting with LF noise and weakly coupled to a HF environment [33]. The problem is characterized by an extremely small gap in the energy spectrum of qubits in the middle of annealing. Solving the problem requires the right combination of Bloch-Redfield and Marcus approaches as well as a proper consideration of the calculation basis, which takes into account the nonstationary effects. The results of the present paper can also be applied to any other open quantum system within the validity range of the approximations.The paper is organized in the following way. Section 2 describes a single-qubit system to provide the necessary intuition before moving to more complicated multiqubit problems. Section 3 formulates the Hamiltonian and introduces important definitions and notations for the system and the bath. Master equations for the probability distribution of the quantum system are derived in section 4. Section 5 presents the relaxation rates as convolution integrals of the Bloch-Redfield and Marcus envelopes. In the section 6 we show that in equilibrium the master equations obey the detailed balance conditions. We also demonstrate that the convolution expression for the relaxation rates turns into the Marcus or to Bloch-Redfield formulas in the corresponding limits. Dissipative dynamics of a 16 qubit system with an extremely small energy gap is considered in section 7. A brief compilation of the commonly encountered notations is presented in appendix A. In other appendixes we provide a detailed derivation of many important formulas.

show abstract

Section: Discussionsupporting

confidence: 54%

“…We are interested in dissipative evolution of a quantum annealer [8,14,15,33,37] treated as a system of N qubits coupled to a heat bath. The qubits are described by the Hamiltonian:…”

Section: The Hamiltonianmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Theory of open quantum dynamics with hybrid noise

Smirnov

Amin

2018

New J. Phys.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Hence T irr is defined as the temperature where the average time for a moment to escape from energy well is equal to the characteristic measurement time. Apparently, demonstrating the T irr properties not only helps us to understand the SG nature, but also promotes the SG applications in fields as diverse as econophysics, biology, or optimization in computer science . However, experimental probes on T irr are still a challenge due to zero overall magnetization in the whole temperature range, and a systematic study of T irr is also lacking.…”

mentioning

confidence: 99%

Spin‐Glass Irreversibility Temperature and Magnetic Stabilization in Ferromagnet/Spin‐Glass Bilayers

2019

Physica Rapid Research Ltrs

View full text Add to dashboard Cite

The authors numerically studied the irreversibility temperature (Tirr) of a spin glass (SG) and how to use a SG to stabilize a ferromagnet. The rate of Tirr/SG‐anisotropic‐field increment can reach ≈121 K T−1. In ferromagnet/SG bilayers, exchange bias (EB) may be enhanced fivefold at 10 K with increasing interfacial coupling (JIF), whereas the EB blocking temperature (TBEB) does not change. However, decreases in SG exchange coupling (JSG) can twofold enhance TBEB with maintaining a table‐like EB. On the contrary, small JIF and large JSG are both destructive to coercivity. Nevertheless, EB field and coercivity are switchable, opening an opportunity to design temperature‐controlled write‐to‐read switches.

show abstract

“…The developments of QML by the D‐Wave system are still at the beginning stage to be used for problems spanning chemistry, materials, and biology . The limit‐connection quantum hardware can hardly solve large‐scale problems although it can achieve the comparable performance to classical ML in small cases.…”

Section: D‐wave Application Developmentmentioning

confidence: 99%

Quantum machine learning with D‐wave quantum computer

Wang

et al. 2019

Quantum Engineering

View full text Add to dashboard Cite

Summary The new era of artificial intelligence (AI) aims to entangle the relationships among models (characterizations), algorithms, and implementations toward the high‐level intelligence with general cognitive ability, strong robustness, and interpretability, which is intractable for machine learning (ML). Quantum computer provides a new computing paradigm for ML. Although universal quantum computers are still in infancy, special‐purpose D‐Wave machine hopefully becomes the breaking point of commercialized quantum computing. The core principle, quantum annealing (QA), enables the quantum system to naturally evolve toward the low‐energy states. D‐Wave's quantum computer has developed some applications of quantum ML based on quantum‐assisted ML algorithms, quantum Boltzmann machine, etc. Additionally, working with CPUs, quantum processing units is likely to advance ML in a quantum‐inspired way. Thus, a new advanced computing architecture, quantum‐classical hybrid approach consisting of QA, classical computing, and brain‐inspired cognitive science, is required to explore its superiority to universal quantum algorithms and classical ML algorithms. It is important to explore hybrid quantum/classical approaches to overcome the defects of ML such as high dependence on training data, low robustness to the noises, and cognitive impairment. The new framework is expected to gradually form a highly effective, accurate, and adaptive intelligent computing architecture for the next generation of AI.

show abstract

Phase transitions in a programmable quantum spin glass simulator

Cited by 258 publications

References 44 publications

Theory of open quantum dynamics with hybrid noise

Theory of open quantum dynamics with hybrid noise

Spin‐Glass Irreversibility Temperature and Magnetic Stabilization in Ferromagnet/Spin‐Glass Bilayers

Quantum machine learning with D‐wave quantum computer

Contact Info

Product

Resources

About