skip to main content

This content will become publicly available on January 7, 2023

Title: Learning a compass spin model with neural network quantum states
Abstract Neural network quantum states provide a novel representation of the many-body states of interacting quantum systems and open up a promising route to solve frustrated quantum spin models that evade other numerical approaches. Yet its capacity to describe complex magnetic orders with large unit cells has not been demonstrated, and its performance in a rugged energy landscape has been questioned. Here we apply restricted Boltzmann machines (RBMs) and stochastic gradient descent to seek the ground states of a compass spin model on the honeycomb lattice, which unifies the Kitaev model, Ising model and the quantum 120° model with a single tuning parameter. We report calculation results on the variational energy, order parameters and correlation functions. The phase diagram obtained is in good agreement with the predictions of tensor network ansatz, demonstrating the capacity of RBMs in learning the ground states of frustrated quantum spin Hamiltonians. The limitations of the calculation are discussed. A few strategies are outlined to address some of the challenges in machine learning frustrated quantum magnets.
Authors:
; ;
Award ID(s):
2011386
Publication Date:
NSF-PAR ID:
10330474
Journal Name:
Journal of Physics: Condensed Matter
Volume:
34
Issue:
12
Page Range or eLocation-ID:
125802
ISSN:
0953-8984
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    Mesoscopic quantum systems exhibit complex many-body quantum phenomena, where interactions between spins and charges give rise to collective modes and topological states. Even simple, non-interacting theories display a rich landscape of energy states—distinct many-particle configurations connected by spin- and energy-dependent transition rates. The ways in which these energy states interact is difficult to characterize or predict, especially in regimes of frustration where many-body effects create a multiply degenerate landscape. Here, we use network science to characterize the complex interconnection patterns of these energy-state transitions. Using an experimentally verified computational model of electronic transport through quantum antidots, we construct networks where nodes represent accessible energy states and edges represent allowed transitions. We find that these networks exhibit Rentian scaling, which is characteristic of efficient transportation systems in computer circuitry, neural circuitry, and human mobility, and can be used to measure the interconnection complexity of a network. We find that the topological complexity of the state transition networks—as measured by Rent’s exponent— correlates with the amount of current flowing through the antidot system. Furthermore, networks corresponding to points of frustration (due, for example, to spin-blockade effects) exhibit an enhanced topological complexity relative to non-frustrated networks. Our results demonstrate that network characterizationsmore »of the abstract topological structure of energy landscapes capture salient properties of quantum transport. More broadly, our approach motivates future efforts to use network science to understand the dynamics and control of complex quantum systems.

    « less
  2. Cyanuric acid is a triazine derivative that has been identified from reactions performed under prebiotic conditions and has been proposed as a prospective precursor of ancestral RNA. For cyanuric acid to have played a key role during the prebiotic era, it would have needed to survive the harsh electromagnetic radiation conditions reaching the Earth’s surface during prebiotic times (≥200 nm). Therefore, the photostability of cyanuric acid would have been crucial for its accumulation during the prebiotic era. To evaluate the putative photostability of cyanuric acid in water, in this contribution, we employed density functional theory (DFT) and its time-dependent variant (TD-DFT) including implicit and explicit solvent effects. The calculations predict that cyanuric acid has an absorption maximum at ca. 160 nm (7.73 eV), with the lowest-energy absorption band extending to ca. 200 nm in an aqueous solution and exhibiting negligible absorption at longer wavelengths. Excitation of cyanuric acid at 160 nm or longer wavelengths leads to the population of S5,6 singlet states, which have ππ* character and large oscillator strengths (0.8). The population reaching the S5,6 states is expected to internally convert to the S1,2 states in an ultrafast time scale. The S1,2 states, which have nπ* character, are predictedmore »to access a conical intersection with the ground state in a nearly barrierless fashion (ca. ≤ 0.13 eV), thus efficiently returning the population to the ground state. Furthermore, based on calculated spin–orbit coupling elements of ca. 6 to 8 cm−1, the calculations predict that intersystem crossing to the triplet manifold should play a minor role in the electronic relaxation of cyanuric acid. We have also calculated the vertical ionization energy of cyanuric acid at 8.2 eV, which predicts that direct one-photon ionization of cyanuric acid should occur at ca. 150 nm. Collectively, the quantum-chemical calculations predict that cyanuric acid would have been highly photostable under the solar radiation conditions reaching the Earth’s surface during the prebiotic era in an aqueous solution. Of relevance to the chemical origin of life and RNA-first theories, these observations lend support to the idea that cyanuric acid could have accumulated in large quantities during the prebiotic era and thus strengthens its candidature as a relevant prebiotic nucleobase.« less
  3. We present a valence transition model for electron- and hole-doped cuprates, within which there occurs a discrete jump in ionicity Cu2+ -> Cu1+ in both families upon doping, at or near optimal doping in the conventionally prepared electron-doped compounds and at the pseudogap phase transition in the hole-doped materials. In thin films of the T' compounds, the valence transition has occurred already in the undoped state. The phenomenology of the valence transition is closely related to that of the neutral-to-ionic transition in mixed-stack organic charge-transfer solids. Doped cuprates have negative charge-transfer gaps, just as rare-earth nickelates and BaBiO3. The unusually high ionization energy of the closed shell Cu1+ ion, taken together with the dopingdriven reduction in three-dimensional Madelung energy and gain in two-dimensional delocalization energy in the negative charge transfer gap state drives the transition in the cuprates. The combined effects of strong correlations and small d-p electron hoppings ensure that the systems behave as effective 1/2-filled Cu band with the closed shell electronically inactive O2- ions in the undoped state, and as correlated two-dimensional geometrically frustrated 1/4-filled oxygen hole band, now with electronically inactive closed-shell Cu1+ ions, in the doped state. The model thus gives microscopic justification for themore »two-fluid models suggested by many authors. The theory gives the simplest yet most comprehensive understanding of experiments in the normal states. The robust commensurate antiferromagnetism in the conventional T' crystals, the strong role of oxygen deficiency in driving superconductivity and charge carrier sign corresponding to holes at optimal doping are all manifestations of the same quantum state. In the hole-doped pseudogapped state, there occurs a biaxial commensurate period 4 charge density wave state consisting of O1- -Cu l(1+)-O1- spin singlets that coexists with broken rotational C-4 symmetry due to intraunit cell oxygen inequivalence. Finite domains of this broken symmetry state will exhibit twodimensional chirality and the polar Kerr effect. Superconductivity within the model results from a destabilization of the 1/4-filled band paired Wigner crystal [Phys. Rev. B 93, 165110 (2016) and ihid. 93, 205111 (2016)]. We posit that a similar valence transition, Ir4+ -> Ir3+, occurs upon electron doping Sr2IrO4. We make testable experimental predictions in cuprates including superoxygenated La2CuO4+delta and iridates. Finally, as indirect evidence for the valence bond theory of superconductivity proposed here, we note that there exist an unusually large number of unconventional superconductors that exhibit superconductivity proximate to exotic charge ordered states, whose band fillings are universally 1/4 or 3/4, exactly where the paired Wigner crystal is most stable.« less
  4. Abstract

    Quantum annealing is a powerful alternative model of quantum computing, which can succeed in the presence of environmental noise even without error correction. However, despite great effort, no conclusive demonstration of a quantum speedup (relative to state of the art classical algorithms) has been shown for these systems, and rigorous theoretical proofs of a quantum advantage (such as the adiabatic formulation of Grover’s search problem) generally rely on exponential precision in at least some aspects of the system, an unphysical resource guaranteed to be scrambled by experimental uncertainties and random noise. In this work, we propose a new variant of quantum annealing, called RFQA, which can maintain a scalable quantum speedup in the face of noise and modest control precision. Specifically, we consider a modification of flux qubit-based quantum annealing which includes low-frequency oscillations in the directions of the transverse field terms as the system evolves. We show that this method produces a quantum speedup for finding ground states in the Grover problem and quantum random energy model, and thus should be widely applicable to other hard optimization problems which can be formulated as quantum spin glasses. Further, we explore three realistic noise channels and show that the speedupmore »from RFQA is resilient to 1/f-like local potential fluctuations and local heating from interaction with a sufficiently low temperature bath. Another noise channel, bath-assisted quantum cooling transitions, actually accelerates the algorithm and may outweigh the negative effects of the others. We also detail how RFQA may be implemented experimentally with current technology.

    « less
  5. The study of low-dimensional quantum systems has proven to be a particularly fertile field for discovering novel types of quantum matter. When studied numerically, low-energy states of low-dimensional quantum systems are often approximated via a tensor-network description. The tensor network's utility in studying short range correlated states in 1D have been thoroughly investigated, with numerous examples where the treatment is essentially exact. Yet, despite the large number of works investigating these networks and their relations to physical models, examples of exact correspondence between the ground state of a quantum critical system and an appropriate scale-invariant tensor network have eluded us so far. Here we show that the features of the quantum-critical Motzkin model can be faithfully captured by an analytic tensor network that exactly represents the ground state of the physical Hamiltonian. In particular, our network offers a two-dimensional representation of this state by a correspondence between walks and a type of tiling of a square lattice. We discuss connections to renormalization and holography.