QLT (Canada)

Quantum state tomography (QST) is the art of reconstructing an unknown quantum state through measurements. It is a key primitive for developing quantum technologies. Neural network quantum state tomography (NNQST), which aims to reconstruct the quantum state via a neural network ansatz, is often implemented via a basis-dependent cross-entropy loss function. State-of-the-art implementations of NNQST are often restricted to characterizing a particular subclass of states, to avoid an exponential growth in the number of required measurement settings. To provide a more broadly applicable method for efficient state reconstruction, we present "neural-shadow quantum state tomography" (NSQST)-an alternative neural network-based QST protocol that uses infidelity as the loss function. The infidelity is estimated using the classical shadows of the target state. Infidelity is a natural choice for training loss, benefiting from the proven measurement sample efficiency of the classical shadow formalism. Furthermore, NSQST is robust against various types of noise without any error mitigation. We numerically demonstrate the advantage of NSQST over NNQST at learning the relative phases of three target quantum states of practical interest, as well as the advantage over direct shadow estimation. NSQST greatly extends the practical reach of NNQST and provides a novel route to effective quantum state tomography.

We introduce two quantum algorithms for solving structured prediction problems. We first show that a stochastic gradient descent that uses the quantum minimum finding algorithm and takes its probabilistic failure into account solves the structured prediction problem with a runtime that scales with the square root of the size of the label space, and in $$\tilde{O}\left (1/\epsilon \right )$$ with respect to the precision, 𝜖, of the solution. Motivated by robust inference techniques in machine learning, we then introduce another quantum algorithm that solves a smooth approximation of the structured prediction problem with a similar quantum speedup in the size of the label space and a similar scaling in the precision parameter. In doing so, we analyze a variant of stochastic gradient descent for convex optimization in the presence of an additive error in the calculation of the gradients, and show that its convergence rate does not deteriorate if the additive errors are of the order $$O(\sqrt{\epsilon})$$ . This algorithm uses quantum Gibbs sampling at temperature Ω(𝜖) as a subroutine. Based on these theoretical observations, we propose a method for using quantum Gibbs samplers to combine feedforward neural networks with probabilistic graphical models for quantum machine learning. Our numerical results using Monte Carlo simulations on an image tagging task demonstrate the benefit of the approach.

Quantum computers have the potential to advance material design and drug discovery by performing costly electronic structure calculations. A critical aspect of this application requires optimizing the limited resources of the quantum hardware. Here, we experimentally demonstrate an end-to-end pipeline that focuses on minimizing quantum resources while maintaining accuracy. Using density matrix embedding theory as a problem decomposition technique, and an ion-trap quantum computer, we simulate a ring of 10 hydrogen atoms without freezing any electrons. The originally 20-qubit system is decomposed into 10 two-qubit problems, making it amenable to currently available hardware. Combining this decomposition with a qubit coupled cluster circuit ansatz, circuit optimization, and density matrix purification, we accurately reproduce the potential energy curve in agreement with the full configuration interaction energy in the minimal basis set. Our experimental results are an early demonstration of the potential for problem decomposition to accurately simulate large molecules on quantum hardware.

Signed graphs serve as a primary tool for modelling social networks. They can represent relationships between individuals (i.e., nodes) with the use of signed edges. Finding communities in a signed graph is of great importance in many areas, for example, targeted advertisement. We propose an algorithm to detect multiple communities in a signed graph. Our method reduces the multi-community detection problem to a quadratic binary unconstrained optimization problem and uses state-of-the-art quantum or classical optimizers to find an optimal assignment of each individual to a specific community.

There have been multiple attempts to demonstrate that quantum annealing and, in particular, quantum annealing on quantum annealing machines, has the potential to outperform current classical optimization algorithms implemented on CMOS technologies. The benchmarking of these devices has been controversial. Initially, random spin-glass problems were used, however, these were quickly shown to be not well suited to detect any quantum speedup. Subsequently, benchmarking shifted to carefully crafted synthetic problems designed to highlight the quantum nature of the hardware while (often) ensuring that classical optimization techniques do not perform well on them. Even worse, to date a true sign of improved scaling with the number of problem variables remains elusive when compared to classical optimization techniques. Here, we analyze the readiness of quantum annealing machines for real-world application problems. These are typically not random and have an underlying structure that is hard to capture in synthetic benchmarks, thus posing unexpected challenges for optimization techniques, both classical and quantum alike. We present a comprehensive computational scaling analysis of fault diagnosis in digital circuits, considering architectures beyond D-wave quantum annealers. We find that the instances generated from real data in multiplier circuits are harder than other representative random spin-glass benchmarks with a comparable number of variables. Although our results show that transverse-field quantum annealing is outperformed by state-of-the-art classical optimization algorithms, these benchmark instances are hard and small in the size of the input, therefore representing the first industrial application ideally suited for testing near-term quantum annealers and other quantum algorithmic strategies for optimization problems.

There have been multiple attempts to design synthetic benchmark problems with the goal of detecting quantum speedup in current quantum annealing (QA) machines. To date, classical heuristics have consistently outperformed quantum annealing based approaches. Here we introduce a class of problems based on frustrated cluster loops—deceptive cluster loops—for which all currently known state of-the-art classical heuristics are outperformed by the DW2000Q QA machine. While there is a sizable constant speedup over all known classical heuristics, a noticeable improvement in the scaling remains elusive. These results represent the first steps towards a detection of potential quantum speedup, albeit without a scaling improvement and for synthetic benchmark problems.

In an effort to overcome the limitations of random spin-glass benchmarks for quantum annealers, focus has shifted to carefully crafted gadget-based problems whose logical structure typically has a planar topology. Recent experiments on these gadget problems using a commercially available quantum annealer have demonstrated an impressive performance over a selection of commonly used classical optimisation heuristics. Here, we show that efficient classical optimisation techniques, such as minimum-weight-perfect matching, can solve these gadget problems exactly and in polynomial time. We present approaches on how to mitigate this shortcoming of commonly used benchmark problems based on planar logical topologies.

The proinflammatory cytokine interleukin-20 (IL-20) may exert the majority of its activity in the skin. We examined the effect of various treatments including several forms of phototherapy on IL-20 expression using cultured normal human epithelial keratinocytes (NHEK). Broadband UVB light, recombinant (r) IL-1 and rIL-8 increased, while hydrocortisone reduced, NHEK supernatant IL-20 levels. Elevation of NHEK IL-20 mRNA and maximal supernatant IL-20 levels occurred with a UVB light dose (40 mJ cm(-2)) that reduced cell viability by approximately 50%. While this UVB light dose also elevated supernatant IL-1 alpha and IL-8 levels, antibody neutralization studies indicated that neither of these cytokines was directly responsible for this increase in IL-20 expression. However, the elevation in IL-20 levels was fully inhibited by the p38 mitogen-activated protein kinase (MAPK) inhibitor SB-203580, suggesting involvement of this stress signaling pathway in this UVB light response. Photodynamic therapy (PDT) with the photosensitizer lemuteporfin, UVA light, cisplatin, lipopolysaccharide (LPS), tumor necrosis factor-alpha (TNF-alpha) or recombinant interferon-gamma (rIFN-gamma) either had little effect or decreased NHEK supernatant IL-20 levels. Reduced IL-20 levels paralleled the cytotoxic actions of PDT, UVA light or cisplatin and the antiproliferative effect of rIFN-gamma. Neither rIL-20 supplementation nor anti-IL-20 antibody treatments affected cell viability indicating that soluble IL-20 did not affect the short-term survival of UVB light-irradiated NHEK. Stimulation of IL-20 expression in keratinocytes by UVB light suggests that this cytokine might participate in skin responses to this ever-present environmental factor and potentially has a role in UV light-associated dermatoses.

Porfimer sodium is an agent used for photodynamic therapy (PDT) of cancer and other pre-malignant conditions such as high grade dysplasia in Barrett's oesophagus. Since it is activated by non-thermal red light after a 2-day time interval to allow distribution in the target tissues, its pharmacokinetic properties are relevant to the timing of light treatment and to the period of protection against photosensitivity reactions. With the recent availability of a reliable assay overcoming the limitations of previous assays, two definitive pharmacokinetic studies were undertaken. To determine if sex or a target disease state (cancer) have an effect on porfimer sodium pharmacokinetic parameters. Twenty-four healthy volunteers (12 men and 12 women) and five male patients with oesophageal cancer undergoing palliative PDT for their obstructive lesions were enrolled. All received an intravenous injection of porfimer sodium (Photofrin) 2 mg/kg over 3-5 minutes and underwent serial blood samplings over 35 days postdose. Porfimer sodium was quantified in serum by a validated spectrofluorometry assay and low-level pre-existing interference was subtracted from postdose concentrations. The two sexes had comparable maximum serum concentrations with a ratio of 0.95. Women tended to have higher areas under the serum concentration-time curve from time zero to the last sampling time, and from time zero to infinity than men, but the difference did not reach significance (ratios of means of 1.18 and 1.20, respectively). Elimination parameters also showed no sex-related differences with a mean distribution half-life of 9.5 hours, clearance of 0.88 mL/h/kg and a terminal elimination half-life of 415 hours (17.3 days). The sexes only differed significantly for the time to reach maximum serum concentration (means of 1.54 and 0.165 hours, for women and men, respectively; p = 0.0239). This is probably because of the sparse sampling schedule and the plateau behaviour of the initial concentrations. The pharmacokinetic parameters in cancer patients were generally comparable to healthy volunteers. However, the mean terminal elimination half-life was 30% shorter (283 hours or 11.8 days) in cancer patients. Sex does not have an effect on porfimer sodium pharmacokinetics. The presence of advanced oesophageal cancer does not seem to have any influence either. These findings confirm that there is no need for sex-specific label recommendations. Also, the elimination phase of porfimer sodium starting progressively from 24 hours postdose supports the recommended time interval for laser light application, the window for PDT debridement and the skin protection period of at least 30 days.