Origin Quantum Computing Technology Company (China)

Accurate prediction of molecular properties is crucial for accelerating the development of new drugs, and quantum machine learning (QML) holds great promise in this domain. A typical QML pipeline comprises two core stages: encoding classical data into quantum representations followed by training and prediction using quantum computing-based machine learning (ML) models. In this article, we focus on the initial encoding stage and propose an effective quantum feature extraction approach for molecular graph data, introducing quantum node embedding and quantum edge embedding methods. We developed a hybrid quantum-classical ML framework and implemented several quantum-embedded graph neural network (QEGNN) models to evaluate the proposed method. Experiments conducted on three benchmark data sets with diverse molecular property prediction tasks demonstrate that QEGNN models consistently achieve higher accuracy, improved stability, and significantly reduced parameter complexity─hallmarks of quantum advantage. Furthermore, we validate the reliability of the quantum embedding approach on the superconducting quantum processor "Wukong," showing that the models retain stable performance even under the constraints of current noisy quantum hardware. This work highlights the potential of QML and paves the way for the development of universal QML models.

The rapid development of quantum computers has enabled demonstrations of quantum advantages on various tasks. However, real quantum systems are always dissipative due to their inevitable interaction with the environment, and the resulting non-unitary dynamics make quantum simulation challenging with only unitary quantum gates. In this work, we present an innovative and scalable method to simulate open quantum systems using quantum computers. We define an adjoint density matrix as a counterpart of the true density matrix, which reduces to a mixed-unitary quantum channel and thus can be effectively sampled using quantum computers. This method has several benefits, including no need for auxiliary qubits and noteworthy scalability. Moreover, some long-time properties like steady states and the thermal equilibrium can also be investigated as the adjoint density matrix and the true dissipated one converge to the same state. Finally, we present deployments of this theory in the dissipative quantum XY model for the evolution of correlation and entropy with short-time dynamics and the disordered Heisenberg model for many-body localization with long-time dynamics. This work promotes the study of real-world many-body dynamics with quantum computers, highlighting the potential to demonstrate practical quantum advantages.

Abstract Synthesized spin-orbit coupling (SSOC) is crucial for the operation of spin qubits in silicon quantum dot, as it address the challenge posed by the inherently weak intrinsic spin-orbit coupling (ISOC) in silicon. Here, we investigate the anisotropic properties of single spin qubit in silicon metal-oxide-semiconductor (Si-MOS) quantum dot and provide experimental evidence for the control of SSOC. Additionally, we experimentally demonstrate that tuning the operating point away from the conventional configuration can enhance the quality factor of the spin qubit. These findings lay a foundation for the realization of high-quality tunable spin-orbit qubits.

Quantum interference is a natural consequence of wave-particle duality in quantum mechanics, and is widely observed at the atomic scale. One interesting manifestation of quantum interference is coherent population trapping (CPT), first proposed in three-level driven atomic systems and observed in quantum optical experiments. Here, we demonstrate CPT in a gate-defined semiconductor double quantum dot (DQD), with some unique twists as compared to the atomic systems. Specifically, we observe CPT in both driven and nondriven situations. We further show that CPT in a driven DQD could be used to generate adiabatic state transfer. Moreover, our experiment reveals a nontrivial modulation to the CPT caused by the longitudinal driving field, yielding an odd-even effect and a tunable CPT. Our results broaden the field of CPT, and open up the possibility of quantum simulation and quantum computation based on adiabatic passage in quantum dot systems.

A challenge in building large-scale superconducting quantum processors is the precise control and manipulation of the qubit state. However, the crosstalk between the microwave control lines impedes the parallel execution of high-fidelity digital and analog quantum operations. Here, we propose and demonstrate a universal compensation protocol for calibrating the microwave signal crosstalk. We also introduce amplified error sequences to optimize the accuracy. Furthermore, we show a definitive improvement in parallel gate operations with crosstalk cancellation, demonstrating the technique's effectiveness. This work paves the way for superconducting hardware that features automated calibration of microwave crosstalk, leading to enhanced fidelities in multiqubit circuits.

We fabricate and characterize a hybrid quantum device that consists of five gate-defined double quantum dots (DQDs) and a high-impedance NbTiN transmission resonator. The controllable interactions between DQDs and the resonator are spectroscopically explored by measuring the microwave transmission through the resonator in the detuning parameter space. Utilizing the high tunability of the system parameters and the high cooperativity (Ctotal > 17.6) interaction between the qubit ensemble and the resonator, we tune the charge-photon coupling and observe the collective microwave response changing from linear to nonlinear. Our results present the maximum number of DQDs coupled to a resonator and manifest a potential platform for scaling up qubits and studying collective quantum effects in semiconductor-superconductor hybrid cavity quantum electrodynamics systems.

Hole spin qubits based on germanium (Ge) have strong tunable spin-orbit interaction (SOI) and ultrafast qubit operation speed. Here we report that the Rabi frequency (fRabi) of a hole spin qubit in a Ge hut wire (HW) double quantum dot (DQD) is electrically tuned through the detuning energy (ϵ) and middle gate voltage (VM). fRabi gradually decreases with increasing ϵ; on the contrary, fRabi is positively correlated with VM. We attribute our results to the change of electric field on SOI and the contribution of the excited state in quantum dots to fRabi. We further demonstrate an ultrafast fRabi exceeding 1.2 GHz, which indicates the strong SOI in our device. The discovery of an ultrafast and electrically tunable fRabi in a hole spin qubit has potential applications in semiconductor quantum computing.

Variational quantum algorithms (VQAs) are widely applied in the noisy intermediate-scale quantum era and are expected to demonstrate quantum advantage. However, training VQAs faces difficulties, one of which is the so-called barren plateaus (BPs) phenomenon, where gradients of cost functions vanish exponentially with the number of qubits. In this paper, inspired by transfer learning, where knowledge of pre-solved tasks could be further used in a different but related work with training efficiency improved, we report a parameter initialization method to mitigate BP. In the method, a small-sized task is solved with a VQA. Then the ansatz and its optimum parameters are transferred to tasks with larger sizes. Numerical simulations show that this method could mitigate BP and improve training efficiency. A brief discussion on how this method can work well is also provided. This work provides a reference for mitigating BP, and therefore, VQAs could be applied to more practical problems.