Advanced quantum technologies最新文献

The controlled integration of quantum dots (QDs) as single-photon emitters into quantum light sources is essential for the implementation of large-scale quantum networks. In this study, we employ the deterministic in situ electron-beam lithography (iEBL) nanotechnology platform to integrate individual QDs with high accuracy and process yield into a circular Bragg grating (CBG) resonators. Notably, CBG devices comprising just 3 to 4 rings exhibit photon extraction efficiencies comparable to those of structures with more rings. This facilitates faster fabrication, reduces the device footprint, and enables compatibility with electrical contacting. To demonstrate the scalability of this process, we present results of 95 optically active QD-CBG devices fabricated across two lithography sessions. These devices exhibit bright, narrow-linewidth single-photon emission with excellent optical quality. To evaluate QD placement accuracy, we apply a powerful characterization technique that combines cathodoluminescence (CL) mapping and scanning electron microscopy. Statistical analysis of these devices reveals that our iEBL approach enables high alignment accuracy and a process yield of over $��>�90�\%�$ across various CBG geometries. Our findings highlight a reliable route toward the scalable fabrication of high-performance QD-based single-photon sources for use in photonic quantum technology applications.

The Hyperband-QNN algorithm achieves adaptive customization of quantum neural network structure by skilfully integrating the essence of neural networks and the Hyperband optimization algorithm, perfectly matching the needs of various specific tasks. This groundbreaking method not only opens up a new path for the design of quantum neural networks but also sets a model for the deep integration of quantum computing and traditional computing. More in article number 2400178, Zheng Shan and co-workers.

Quantum-noise-driven diffusion models are proposed here as a novel class of quantum generative AI algorithms. These models aim to exploit the intrinsic noise of currently available quantum processing units, not as an issue to be solved by quantum error mitigation and correction, but instead as a beneficial resource to generate artificial, classical or quantum, data sampled from some unknown and usually very complex probability distribution that could be difficult or even impossible to sample from via classical computers. More in article number 2300401, Filippo Caruso and co-workers.

In article number 2400603, Alexey Melnikov and co-workers present a method to enhance the generalization capability of quantum machine learning models through controllable noise regularization. The cover visualizes a quantum circuit in which quantum gates are interleaved with engineered noise channels of tunable strength λ. This structured injection of noise acts as an implicit regularizer, stabilizing the training process and improving the model's predictive performance on previously unseen data.

This image illustrates the process of Rydberg-based quantum-enhanced simulated annealing (QESA), which achieves faster computational performance than classical simulated annealing (SA). Harnessing Rydberg interactions tailored to the maximum independent set problem, QESA demonstrates a clear quantum advantage in heuristic optimization, pointing to a promising route for tackling complex combinatorial problems more efficiently than classical methods. More in article number 2500070, Jaewook Ahn and co-workers.

A large-scale quantum computer promises quantum algorithms with an exponential speed increase over classical counterparts. Ion implanted donor atom qubits in isotopically enriched $��{}^{28}�\mathrm{Si}�$ have demonstrated leading coherence times and gate fidelities. Large-scale donor architectures must be tolerant to positional uncertainties owing to ion straggling consistent with the implant energy required to place near surface donor atoms allowing manipulation by surface control electrodes. These issues are addressed with a 40 million silicon atom quantum model based on the atomistic tight-binding technique in the context of the flip-flop qubit architecture. The donor electron wavefunction operates in a state of equal superposition between the buried host donor atom and below the surface gate oxide. Key issues include: donor positional uncertainty from ion straggling coupled with non-uniform control gate potentials, the coupling of donor wave functions to gate-defined quantum wells, and the effect of bystander donors including the mitigation of non-functional adventitious channeled ions by re-implantation employing deterministic ion implantation. An upper constraint of 9 keV $��{}^{31}�P�$ ions minimizes straggling allowing quantum superposition and is compatible with existing fabrication techniques. With heavy donors, such as $��{}^{123}�\mathrm{Sb}�$, the constraints are relaxed because, for a given implant depth, the heavy donors straggle less than light donors.

In the current quantum computing paradigm, significant focus is placed on the reduction or mitigation of quantum decoherence. When designing new quantum processing units, the general objective is to reduce the amount of noise qubits are subject to, and in algorithm design, a large effort is underway to provide scalable error correction or mitigation techniques. Yet some previous work has indicated that certain classes of quantum algorithms, such as quantum machine learning, may, in fact, be intrinsically robust to or even benefit from the presence of a small amount of noise. Here, we demonstrate that noise levels in quantum hardware can be effectively tuned to enhance the ability of quantum neural networks to generalize data, acting akin to regularisation in classical neural networks. As an example, we consider two regression tasks, where, by tuning the noise level in the circuit, we demonstrated improvement of the validation mean squared error loss. Moreover, we demonstrate the method's effectiveness by numerically simulating QNN training on a realistic model of a noisy superconducting quantum computer.

Bose-Einstein condensates (BECs) provide an ideal platform for exploring novel quantum states (QTs) in synthetic spin-orbit coupling (SOC). In article number 2500431, Yun Liu and Zu-Jian Ying analyse the interplay of SOC with other interactions and potential geometry by BECs in concentric annular traps. Various exotic QTs emerge, including facial-makeup states, fissure states, stripe states, half-disk states, half-skyrmion fence, etc. Peculiar density-phase separation is noticed. The study illustrates manipulations of exotic QTs and supplies abundant quantum resources for potential applications.